Hu, Zixi; Yao, Zhewei; Li, Jinglai
2015-01-01
Many scientific and engineering problems require to perform Bayesian inferences for unknowns of infinite dimension. In such problems, many standard Markov Chain Monte Carlo (MCMC) algorithms become arbitrary slow under the mesh refinement, which is referred to as being dimension dependent. To this end, a family of dimensional independent MCMC algorithms, known as the preconditioned Crank-Nicolson (pCN) methods, were proposed to sample the infinite dimensional parameters. In this work we devel...
Allen Rodrigo
2006-01-01
Full Text Available Using the structured serial coalescent with Bayesian MCMC and serial samples, we estimate population size when some demes are not sampled or are hidden, ie ghost demes. It is found that even with the presence of a ghost deme, accurate inference was possible if the parameters are estimated with the true model. However with an incorrect model, estimates were biased and can be positively misleading. We extend these results to the case where there are sequences from the ghost at the last time sample. This case can arise in HIV patients, when some tissue samples and viral sequences only become available after death. When some sequences from the ghost deme are available at the last sampling time, estimation bias is reduced and accurate estimation of parameters associated with the ghost deme is possible despite sampling bias. Migration rates for this case are also shown to be good estimates when migration values are low.
Bayesian Optimization for Adaptive MCMC
Mahendran, Nimalan; Wang, Ziyu; Hamze, Firas; De Freitas, Nando
2011-01-01
This paper proposes a new randomized strategy for adaptive MCMC using Bayesian optimization. This approach applies to non-differentiable objective functions and trades off exploration and exploitation to reduce the number of potentially costly objective function evaluations. We demonstrate the strategy in the complex setting of sampling from constrained, discrete and densely connected probabilistic graphical models where, for each variation of the problem, one needs to adjust the parameters o...
Measuring the reliability of MCMC inference with bidirectional Monte Carlo
Grosse, Roger B.; Ancha, Siddharth; Roy, Daniel M.
2016-01-01
Markov chain Monte Carlo (MCMC) is one of the main workhorses of probabilistic inference, but it is notoriously hard to measure the quality of approximate posterior samples. This challenge is particularly salient in black box inference methods, which can hide details and obscure inference failures. In this work, we extend the recently introduced bidirectional Monte Carlo technique to evaluate MCMC-based posterior inference algorithms. By running annealed importance sampling (AIS) chains both ...
Complexity analysis of accelerated MCMC methods for Bayesian inversion
Hoang, Viet Ha; Schwab, Christoph; Stuart, Andrew M.
2013-08-01
The Bayesian approach to inverse problems, in which the posterior probability distribution on an unknown field is sampled for the purposes of computing posterior expectations of quantities of interest, is starting to become computationally feasible for partial differential equation (PDE) inverse problems. Balancing the sources of error arising from finite-dimensional approximation of the unknown field, the PDE forward solution map and the sampling of the probability space under the posterior distribution are essential for the design of efficient computational Bayesian methods for PDE inverse problems. We study Bayesian inversion for a model elliptic PDE with an unknown diffusion coefficient. We provide complexity analyses of several Markov chain Monte Carlo (MCMC) methods for the efficient numerical evaluation of expectations under the Bayesian posterior distribution, given data δ. Particular attention is given to bounds on the overall work required to achieve a prescribed error level ε. Specifically, we first bound the computational complexity of ‘plain’ MCMC, based on combining MCMC sampling with linear complexity multi-level solvers for elliptic PDE. Our (new) work versus accuracy bounds show that the complexity of this approach can be quite prohibitive. Two strategies for reducing the computational complexity are then proposed and analyzed: first, a sparse, parametric and deterministic generalized polynomial chaos (gpc) ‘surrogate’ representation of the forward response map of the PDE over the entire parameter space, and, second, a novel multi-level Markov chain Monte Carlo strategy which utilizes sampling from a multi-level discretization of the posterior and the forward PDE. For both of these strategies, we derive asymptotic bounds on work versus accuracy, and hence asymptotic bounds on the computational complexity of the algorithms. In particular, we provide sufficient conditions on the regularity of the unknown coefficients of the PDE and on the
boa: An R Package for MCMC Output Convergence Assessment and Posterior Inference
Brian J. Smith
2007-10-01
Full Text Available Markov chain Monte Carlo (MCMC is the most widely used method of estimating joint posterior distributions in Bayesian analysis. The idea of MCMC is to iteratively produce parameter values that are representative samples from the joint posterior. Unlike frequentist analysis where iterative model fitting routines are monitored for convergence to a single point, MCMC output is monitored for convergence to a distribution. Thus, specialized diagnostic tools are needed in the Bayesian setting. To this end, the R package boa was created. This manuscript presents the user's manual for boa, which outlines the use of and methodology upon which the software is based. Included is a description of the menu system, data management capabilities, and statistical/graphical methods for convergence assessment and posterior inference. Throughout the manual, a linear regression example is used to illustrate the software.
Bayesian inference for Hawkes processes
Rasmussen, Jakob Gulddahl
The Hawkes process is a practically and theoretically important class of point processes, but parameter-estimation for such a process can pose various problems. In this paper we explore and compare two approaches to Bayesian inference. The first approach is based on the so-called conditional...... intensity function, while the second approach is based on an underlying clustering and branching structure in the Hawkes process. For practical use, MCMC (Markov chain Monte Carlo) methods are employed. The two approaches are compared numerically using three examples of the Hawkes process....
Bayesian inference for Hawkes processes
Rasmussen, Jakob Gulddahl
2013-01-01
The Hawkes process is a practically and theoretically important class of point processes, but parameter-estimation for such a process can pose various problems. In this paper we explore and compare two approaches to Bayesian inference. The first approach is based on the so-called conditional...... intensity function, while the second approach is based on an underlying clustering and branching structure in the Hawkes process. For practical use, MCMC (Markov chain Monte Carlo) methods are employed. The two approaches are compared numerically using three examples of the Hawkes process....
Frühwirth-Schnatter, Sylvia
1990-01-01
In the paper at hand we apply it to Bayesian statistics to obtain "Fuzzy Bayesian Inference". In the subsequent sections we will discuss a fuzzy valued likelihood function, Bayes' theorem for both fuzzy data and fuzzy priors, a fuzzy Bayes' estimator, fuzzy predictive densities and distributions, and fuzzy H.P.D .-Regions. (author's abstract)
Computationally efficient Bayesian inference for inverse problems.
Marzouk, Youssef M.; Najm, Habib N.; Rahn, Larry A.
2007-10-01
Bayesian statistics provides a foundation for inference from noisy and incomplete data, a natural mechanism for regularization in the form of prior information, and a quantitative assessment of uncertainty in the inferred results. Inverse problems - representing indirect estimation of model parameters, inputs, or structural components - can be fruitfully cast in this framework. Complex and computationally intensive forward models arising in physical applications, however, can render a Bayesian approach prohibitive. This difficulty is compounded by high-dimensional model spaces, as when the unknown is a spatiotemporal field. We present new algorithmic developments for Bayesian inference in this context, showing strong connections with the forward propagation of uncertainty. In particular, we introduce a stochastic spectral formulation that dramatically accelerates the Bayesian solution of inverse problems via rapid evaluation of a surrogate posterior. We also explore dimensionality reduction for the inference of spatiotemporal fields, using truncated spectral representations of Gaussian process priors. These new approaches are demonstrated on scalar transport problems arising in contaminant source inversion and in the inference of inhomogeneous material or transport properties. We also present a Bayesian framework for parameter estimation in stochastic models, where intrinsic stochasticity may be intermingled with observational noise. Evaluation of a likelihood function may not be analytically tractable in these cases, and thus several alternative Markov chain Monte Carlo (MCMC) schemes, operating on the product space of the observations and the parameters, are introduced.
Improving the structure MCMC sampler for Bayesian networks by introducing a new edge reversal move
Grzegorczyk, Marco; Husmeier, Dirk
2008-01-01
Applications of Bayesian networks in systems biology are computationally demanding due to the large number of model parameters. Conventional MCMC schemes based on proposal moves in structure space tend to be too slow in mixing and convergence, and have recently been superseded by proposal moves in t
Inference algorithms and learning theory for Bayesian sparse factor analysis
Bayesian sparse factor analysis has many applications; for example, it has been applied to the problem of inferring a sparse regulatory network from gene expression data. We describe a number of inference algorithms for Bayesian sparse factor analysis using a slab and spike mixture prior. These include well-established Markov chain Monte Carlo (MCMC) and variational Bayes (VB) algorithms as well as a novel hybrid of VB and Expectation Propagation (EP). For the case of a single latent factor we derive a theory for learning performance using the replica method. We compare the MCMC and VB/EP algorithm results with simulated data to the theoretical prediction. The results for MCMC agree closely with the theory as expected. Results for VB/EP are slightly sub-optimal but show that the new algorithm is effective for sparse inference. In large-scale problems MCMC is infeasible due to computational limitations and the VB/EP algorithm then provides a very useful computationally efficient alternative.
BigFoot: Bayesian alignment and phylogenetic footprinting with MCMC
Miklós István
2009-08-01
Full Text Available Abstract Background We have previously combined statistical alignment and phylogenetic footprinting to detect conserved functional elements without assuming a fixed alignment. Considering a probability-weighted distribution of alignments removes sensitivity to alignment errors, properly accommodates regions of alignment uncertainty, and increases the accuracy of functional element prediction. Our method utilized standard dynamic programming hidden markov model algorithms to analyze up to four sequences. Results We present a novel approach, implemented in the software package BigFoot, for performing phylogenetic footprinting on greater numbers of sequences. We have developed a Markov chain Monte Carlo (MCMC approach which samples both sequence alignments and locations of slowly evolving regions. We implement our method as an extension of the existing StatAlign software package and test it on well-annotated regions controlling the expression of the even-skipped gene in Drosophila and the α-globin gene in vertebrates. The results exhibit how adding additional sequences to the analysis has the potential to improve the accuracy of functional predictions, and demonstrate how BigFoot outperforms existing alignment-based phylogenetic footprinting techniques. Conclusion BigFoot extends a combined alignment and phylogenetic footprinting approach to analyze larger amounts of sequence data using MCMC. Our approach is robust to alignment error and uncertainty and can be applied to a variety of biological datasets. The source code and documentation are publicly available for download from http://www.stats.ox.ac.uk/~satija/BigFoot/
Analysis of KATRIN data using Bayesian inference
Riis, Anna Sejersen; Weinheimer, Christian
2011-01-01
The KATRIN (KArlsruhe TRItium Neutrino) experiment will be analyzing the tritium beta-spectrum to determine the mass of the neutrino with a sensitivity of 0.2 eV (90% C.L.). This approach to a measurement of the absolute value of the neutrino mass relies only on the principle of energy conservation and can in some sense be called model-independent as compared to cosmology and neutrino-less double beta decay. However by model independent we only mean in case of the minimal extension of the standard model. One should therefore also analyse the data for non-standard couplings to e.g. righthanded or sterile neutrinos. As an alternative to the frequentist minimization methods used in the analysis of the earlier experiments in Mainz and Troitsk we have been investigating Markov Chain Monte Carlo (MCMC) methods which are very well suited for probing multi-parameter spaces. We found that implementing the KATRIN chi squared function in the COSMOMC package - an MCMC code using Bayesian parameter inference - solved the ...
Bayesian-MCMC-based parameter estimation of stealth aircraft RCS models
Xia, Wei; Dai, Xiao-Xia; Feng, Yuan
2015-12-01
When modeling a stealth aircraft with low RCS (Radar Cross Section), conventional parameter estimation methods may cause a deviation from the actual distribution, owing to the fact that the characteristic parameters are estimated via directly calculating the statistics of RCS. The Bayesian-Markov Chain Monte Carlo (Bayesian-MCMC) method is introduced herein to estimate the parameters so as to improve the fitting accuracies of fluctuation models. The parameter estimations of the lognormal and the Legendre polynomial models are reformulated in the Bayesian framework. The MCMC algorithm is then adopted to calculate the parameter estimates. Numerical results show that the distribution curves obtained by the proposed method exhibit improved consistence with the actual ones, compared with those fitted by the conventional method. The fitting accuracy could be improved by no less than 25% for both fluctuation models, which implies that the Bayesian-MCMC method might be a good candidate among the optimal parameter estimation methods for stealth aircraft RCS models. Project supported by the National Natural Science Foundation of China (Grant No. 61101173), the National Basic Research Program of China (Grant No. 613206), the National High Technology Research and Development Program of China (Grant No. 2012AA01A308), the State Scholarship Fund by the China Scholarship Council (CSC), and the Oversea Academic Training Funds, and University of Electronic Science and Technology of China (UESTC).
Roche, Alexis
2015-01-01
This paper revisits the concept of composite likelihood from the perspective of probabilistic inference, and proposes a generalization called super composite likelihood for sharper inference in multiclass problems. It is argued that, beside providing a new interpretation and a general justification of na\\"ive Bayes procedures, super composite likelihood yields a much wider class of discriminative models suitable for unsupervised and weakly supervised learning.
Bayesian inference tools for inverse problems
Mohammad-Djafari, Ali
2013-08-01
In this paper, first the basics of Bayesian inference with a parametric model of the data is presented. Then, the needed extensions are given when dealing with inverse problems and in particular the linear models such as Deconvolution or image reconstruction in Computed Tomography (CT). The main point to discuss then is the prior modeling of signals and images. A classification of these priors is presented, first in separable and Markovien models and then in simple or hierarchical with hidden variables. For practical applications, we need also to consider the estimation of the hyper parameters. Finally, we see that we have to infer simultaneously on the unknowns, the hidden variables and the hyper parameters. Very often, the expression of this joint posterior law is too complex to be handled directly. Indeed, rarely we can obtain analytical solutions to any point estimators such the Maximum A posteriori (MAP) or Posterior Mean (PM). Three main tools are then can be used: Laplace approximation (LAP), Markov Chain Monte Carlo (MCMC) and Bayesian Variational Approximations (BVA). To illustrate all these aspects, we will consider a deconvolution problem where we know that the input signal is sparse and propose to use a Student-t prior for that. Then, to handle the Bayesian computations with this model, we use the property of Student-t which is modelling it via an infinite mixture of Gaussians, introducing thus hidden variables which are the variances. Then, the expression of the joint posterior of the input signal samples, the hidden variables (which are here the inverse variances of those samples) and the hyper-parameters of the problem (for example the variance of the noise) is given. From this point, we will present the joint maximization by alternate optimization and the three possible approximation methods. Finally, the proposed methodology is applied in different applications such as mass spectrometry, spectrum estimation of quasi periodic biological signals and
Probability biases as Bayesian inference
Andre; C. R. Martins
2006-11-01
Full Text Available In this article, I will show how several observed biases in human probabilistic reasoning can be partially explained as good heuristics for making inferences in an environment where probabilities have uncertainties associated to them. Previous results show that the weight functions and the observed violations of coalescing and stochastic dominance can be understood from a Bayesian point of view. We will review those results and see that Bayesian methods should also be used as part of the explanation behind other known biases. That means that, although the observed errors are still errors under the be understood as adaptations to the solution of real life problems. Heuristics that allow fast evaluations and mimic a Bayesian inference would be an evolutionary advantage, since they would give us an efficient way of making decisions. %XX In that sense, it should be no surprise that humans reason with % probability as it has been observed.
Numerical approximations for speeding up mcmc inference in the infinite relational model
Schmidt, Mikkel Nørgaard; Albers, Kristoffer Jon
The infinite relational model (IRM) is a powerful model for discovering clusters in complex networks; however, the computational speed of Markov chain Monte Carlo inference in the model can be a limiting factor when analyzing large networks. We investigate how using numerical approximations of th...... on test data. We demonstrate that the computational time for MCMC inference can be reduced by a factor of two without affecting the performance, making it worthwhile in practical situations when on a computational budget....
Bayesian Inference for Radio Observations
Lochner, Michelle; Zwart, Jonathan T L; Smirnov, Oleg; Bassett, Bruce A; Oozeer, Nadeem; Kunz, Martin
2015-01-01
(Abridged) New telescopes like the Square Kilometre Array (SKA) will push into a new sensitivity regime and expose systematics, such as direction-dependent effects, that could previously be ignored. Current methods for handling such systematics rely on alternating best estimates of instrumental calibration and models of the underlying sky, which can lead to inaccurate uncertainty estimates and biased results because such methods ignore any correlations between parameters. These deconvolution algorithms produce a single image that is assumed to be a true representation of the sky, when in fact it is just one realisation of an infinite ensemble of images compatible with the noise in the data. In contrast, here we report a Bayesian formalism that simultaneously infers both systematics and science. Our technique, Bayesian Inference for Radio Observations (BIRO), determines all parameters directly from the raw data, bypassing image-making entirely, by sampling from the joint posterior probability distribution. Thi...
Bayesian inference on proportional elections.
Brunello, Gabriel Hideki Vatanabe; Nakano, Eduardo Yoshio
2015-01-01
Polls for majoritarian voting systems usually show estimates of the percentage of votes for each candidate. However, proportional vote systems do not necessarily guarantee the candidate with the most percentage of votes will be elected. Thus, traditional methods used in majoritarian elections cannot be applied on proportional elections. In this context, the purpose of this paper was to perform a Bayesian inference on proportional elections considering the Brazilian system of seats distribution. More specifically, a methodology to answer the probability that a given party will have representation on the chamber of deputies was developed. Inferences were made on a Bayesian scenario using the Monte Carlo simulation technique, and the developed methodology was applied on data from the Brazilian elections for Members of the Legislative Assembly and Federal Chamber of Deputies in 2010. A performance rate was also presented to evaluate the efficiency of the methodology. Calculations and simulations were carried out using the free R statistical software. PMID:25786259
Skill Rating by Bayesian Inference
Di Fatta, Giuseppe; Haworth, Guy McCrossan; Regan, Kenneth W.
2009-01-01
Systems Engineering often involves computer modelling the behaviour of proposed systems and their components. Where a component is human, fallibility must be modelled by a stochastic agent. The identification of a model of decision-making over quantifiable options is investigated using the game-domain of Chess. Bayesian methods are used to infer the distribution of players’ skill levels from the moves they play rather than from their competitive results. The approach is used on large sets of ...
Inference in hybrid Bayesian networks
Lanseth, Helge; Nielsen, Thomas Dyhre; Rumí, Rafael;
2009-01-01
and reliability block diagrams). However, limitations in the BNs' calculation engine have prevented BNs from becoming equally popular for domains containing mixtures of both discrete and continuous variables (so-called hybrid domains). In this paper we focus on these difficulties, and summarize some of the last...... decade's research on inference in hybrid Bayesian networks. The discussions are linked to an example model for estimating human reliability....
Picchini, Umberto; Forman, Julie Lyng
2016-01-01
applications. A simulation study is conducted to compare our strategy with exact Bayesian inference, the latter resulting two orders of magnitude slower than ABC-MCMC for the considered set-up. Finally, the ABC algorithm is applied to a large size protein data. The suggested methodology is fairly general and...
New inference strategies for solving Markov Decision Processes using reversible jump MCMC
Hoffman, Matthias; Kueck, Hendrik; De Freitas, Nando; Doucet, Arnaud
2012-01-01
In this paper we build on previous work which uses inferences techniques, in particular Markov Chain Monte Carlo (MCMC) methods, to solve parameterized control problems. We propose a number of modifications in order to make this approach more practical in general, higher-dimensional spaces. We first introduce a new target distribution which is able to incorporate more reward information from sampled trajectories. We also show how to break strong correlations between the policy parameters and ...
Quantum Inference on Bayesian Networks
Yoder, Theodore; Low, Guang Hao; Chuang, Isaac
2014-03-01
Because quantum physics is naturally probabilistic, it seems reasonable to expect physical systems to describe probabilities and their evolution in a natural fashion. Here, we use quantum computation to speedup sampling from a graphical probability model, the Bayesian network. A specialization of this sampling problem is approximate Bayesian inference, where the distribution on query variables is sampled given the values e of evidence variables. Inference is a key part of modern machine learning and artificial intelligence tasks, but is known to be NP-hard. Classically, a single unbiased sample is obtained from a Bayesian network on n variables with at most m parents per node in time (nmP(e) - 1 / 2) , depending critically on P(e) , the probability the evidence might occur in the first place. However, by implementing a quantum version of rejection sampling, we obtain a square-root speedup, taking (n2m P(e) -1/2) time per sample. The speedup is the result of amplitude amplification, which is proving to be broadly applicable in sampling and machine learning tasks. In particular, we provide an explicit and efficient circuit construction that implements the algorithm without the need for oracle access.
Nonparametric Bayesian inference in biostatistics
Müller, Peter
2015-01-01
As chapters in this book demonstrate, BNP has important uses in clinical sciences and inference for issues like unknown partitions in genomics. Nonparametric Bayesian approaches (BNP) play an ever expanding role in biostatistical inference from use in proteomics to clinical trials. Many research problems involve an abundance of data and require flexible and complex probability models beyond the traditional parametric approaches. As this book's expert contributors show, BNP approaches can be the answer. Survival Analysis, in particular survival regression, has traditionally used BNP, but BNP's potential is now very broad. This applies to important tasks like arrangement of patients into clinically meaningful subpopulations and segmenting the genome into functionally distinct regions. This book is designed to both review and introduce application areas for BNP. While existing books provide theoretical foundations, this book connects theory to practice through engaging examples and research questions. Chapters c...
New inference strategies for solving Markov Decision Processes using reversible jump MCMC
Hoffman, Matthias; de Freitas, Nando; Doucet, Arnaud
2012-01-01
In this paper we build on previous work which uses inferences techniques, in particular Markov Chain Monte Carlo (MCMC) methods, to solve parameterized control problems. We propose a number of modifications in order to make this approach more practical in general, higher-dimensional spaces. We first introduce a new target distribution which is able to incorporate more reward information from sampled trajectories. We also show how to break strong correlations between the policy parameters and sampled trajectories in order to sample more freely. Finally, we show how to incorporate these techniques in a principled manner to obtain estimates of the optimal policy.
Compiling Relational Bayesian Networks for Exact Inference
Jaeger, Manfred; Chavira, Mark; Darwiche, Adnan
2004-01-01
We describe a system for exact inference with relational Bayesian networks as defined in the publicly available \\primula\\ tool. The system is based on compiling propositional instances of relational Bayesian networks into arithmetic circuits and then performing online inference by evaluating and ...
Compiling Relational Bayesian Networks for Exact Inference
Jaeger, Manfred; Darwiche, Adnan; Chavira, Mark
We describe in this paper a system for exact inference with relational Bayesian networks as defined in the publicly available PRIMULA tool. The system is based on compiling propositional instances of relational Bayesian networks into arithmetic circuits and then performing online inference by eva...
Parallel local approximation MCMC for expensive models
Conrad, Patrick; Davis, Andrew; Marzouk, Youssef; Pillai, Natesh; Smith, Aaron
2016-01-01
Performing Bayesian inference via Markov chain Monte Carlo (MCMC) can be exceedingly expensive when posterior evaluations invoke the evaluation of a computationally expensive model, such as a system of partial differential equations. In recent work [Conrad et al. JASA 2015, arXiv:1402.1694] we described a framework for constructing and refining local approximations of such models during an MCMC simulation. These posterior--adapted approximations harness regularity of the model to reduce the c...
Fast MCMC sampling for Markov jump processes and continuous time Bayesian networks
Rao, Vinayak
2012-01-01
Markov jump processes and continuous time Bayesian networks are important classes of continuous time dynamical systems. In this paper, we tackle the problem of inferring unobserved paths in these models by introducing a fast auxiliary variable Gibbs sampler. Our approach is based on the idea of uniformization, and sets up a Markov chain over paths by sampling a finite set of virtual jump times and then running a standard hidden Markov model forward filtering-backward sampling algorithm over states at the set of extant and virtual jump times. We demonstrate significant computational benefits over a state-of-the-art Gibbs sampler on a number of continuous time Bayesian networks.
Bayesian Inference Methods for Sparse Channel Estimation
Pedersen, Niels Lovmand
2013-01-01
inference algorithms based on the proposed prior representation for sparse channel estimation in orthogonal frequency-division multiplexing receivers. The inference algorithms, which are mainly obtained from variational Bayesian methods, exploit the underlying sparse structure of wireless channel responses......This thesis deals with sparse Bayesian learning (SBL) with application to radio channel estimation. As opposed to the classical approach for sparse signal representation, we focus on the problem of inferring complex signals. Our investigations within SBL constitute the basis for the development of...... Bayesian inference algorithms for sparse channel estimation. Sparse inference methods aim at finding the sparse representation of a signal given in some overcomplete dictionary of basis vectors. Within this context, one of our main contributions to the field of SBL is a hierarchical representation of...
Bayesian Inference on Gravitational Waves
Asad Ali
2015-12-01
Full Text Available The Bayesian approach is increasingly becoming popular among the astrophysics data analysis communities. However, the Pakistan statistics communities are unaware of this fertile interaction between the two disciplines. Bayesian methods have been in use to address astronomical problems since the very birth of the Bayes probability in eighteenth century. Today the Bayesian methods for the detection and parameter estimation of gravitational waves have solid theoretical grounds with a strong promise for the realistic applications. This article aims to introduce the Pakistan statistics communities to the applications of Bayesian Monte Carlo methods in the analysis of gravitational wave data with an overview of the Bayesian signal detection and estimation methods and demonstration by a couple of simplified examples.
AGNfitter: A Bayesian MCMC approach to fitting spectral energy distributions of AGN
Rivera, Gabriela Calistro; Hennawi, Joseph F; Hogg, David W
2016-01-01
We present AGNfitter, a publicly available open-source algorithm implementing a fully Bayesian Markov Chain Monte Carlo method to fit the spectral energy distributions (SEDs) of active galactic nuclei (AGN) from the sub-mm to the UV, allowing one to robustly disentangle the physical processes responsible for their emission. AGNfitter makes use of a large library of theoretical, empirical, and semi-empirical models to characterize both the nuclear and host galaxy emission simultaneously. The model consists of four physical emission components: an accretion disk, a torus of AGN heated dust, stellar populations, and cold dust in star forming regions. AGNfitter determines the posterior distributions of numerous parameters that govern the physics of AGN with a fully Bayesian treatment of errors and parameter degeneracies, allowing one to infer integrated luminosities, dust attenuation parameters, stellar masses, and star formation rates. We tested AGNfitter's performace on real data by fitting the SEDs of a sample...
Bayesian inference for Markov jump processes with informative observations.
Golightly, Andrew; Wilkinson, Darren J
2015-04-01
In this paper we consider the problem of parameter inference for Markov jump process (MJP) representations of stochastic kinetic models. Since transition probabilities are intractable for most processes of interest yet forward simulation is straightforward, Bayesian inference typically proceeds through computationally intensive methods such as (particle) MCMC. Such methods ostensibly require the ability to simulate trajectories from the conditioned jump process. When observations are highly informative, use of the forward simulator is likely to be inefficient and may even preclude an exact (simulation based) analysis. We therefore propose three methods for improving the efficiency of simulating conditioned jump processes. A conditioned hazard is derived based on an approximation to the jump process, and used to generate end-point conditioned trajectories for use inside an importance sampling algorithm. We also adapt a recently proposed sequential Monte Carlo scheme to our problem. Essentially, trajectories are reweighted at a set of intermediate time points, with more weight assigned to trajectories that are consistent with the next observation. We consider two implementations of this approach, based on two continuous approximations of the MJP. We compare these constructs for a simple tractable jump process before using them to perform inference for a Lotka-Volterra system. The best performing construct is used to infer the parameters governing a simple model of motility regulation in Bacillus subtilis. PMID:25720091
An Intuitive Dashboard for Bayesian Network Inference
Reddy, Vikas; Charisse Farr, Anna; Wu, Paul; Mengersen, Kerrie; Yarlagadda, Prasad K. D. V.
2014-03-01
Current Bayesian network software packages provide good graphical interface for users who design and develop Bayesian networks for various applications. However, the intended end-users of these networks may not necessarily find such an interface appealing and at times it could be overwhelming, particularly when the number of nodes in the network is large. To circumvent this problem, this paper presents an intuitive dashboard, which provides an additional layer of abstraction, enabling the end-users to easily perform inferences over the Bayesian networks. Unlike most software packages, which display the nodes and arcs of the network, the developed tool organises the nodes based on the cause-and-effect relationship, making the user-interaction more intuitive and friendly. In addition to performing various types of inferences, the users can conveniently use the tool to verify the behaviour of the developed Bayesian network. The tool has been developed using QT and SMILE libraries in C++.
An Intuitive Dashboard for Bayesian Network Inference
Current Bayesian network software packages provide good graphical interface for users who design and develop Bayesian networks for various applications. However, the intended end-users of these networks may not necessarily find such an interface appealing and at times it could be overwhelming, particularly when the number of nodes in the network is large. To circumvent this problem, this paper presents an intuitive dashboard, which provides an additional layer of abstraction, enabling the end-users to easily perform inferences over the Bayesian networks. Unlike most software packages, which display the nodes and arcs of the network, the developed tool organises the nodes based on the cause-and-effect relationship, making the user-interaction more intuitive and friendly. In addition to performing various types of inferences, the users can conveniently use the tool to verify the behaviour of the developed Bayesian network. The tool has been developed using QT and SMILE libraries in C++
Kernel Bayesian Inference with Posterior Regularization
Song, Yang; Jun ZHU; Ren, Yong
2016-01-01
We propose a vector-valued regression problem whose solution is equivalent to the reproducing kernel Hilbert space (RKHS) embedding of the Bayesian posterior distribution. This equivalence provides a new understanding of kernel Bayesian inference. Moreover, the optimization problem induces a new regularization for the posterior embedding estimator, which is faster and has comparable performance to the squared regularization in kernel Bayes' rule. This regularization coincides with a former th...
Tactile length contraction as Bayesian inference.
Tong, Jonathan; Ngo, Vy; Goldreich, Daniel
2016-08-01
To perceive, the brain must interpret stimulus-evoked neural activity. This is challenging: The stochastic nature of the neural response renders its interpretation inherently uncertain. Perception would be optimized if the brain used Bayesian inference to interpret inputs in light of expectations derived from experience. Bayesian inference would improve perception on average but cause illusions when stimuli violate expectation. Intriguingly, tactile, auditory, and visual perception are all prone to length contraction illusions, characterized by the dramatic underestimation of the distance between punctate stimuli delivered in rapid succession; the origin of these illusions has been mysterious. We previously proposed that length contraction illusions occur because the brain interprets punctate stimulus sequences using Bayesian inference with a low-velocity expectation. A novel prediction of our Bayesian observer model is that length contraction should intensify if stimuli are made more difficult to localize. Here we report a tactile psychophysical study that tested this prediction. Twenty humans compared two distances on the forearm: a fixed reference distance defined by two taps with 1-s temporal separation and an adjustable comparison distance defined by two taps with temporal separation t ≤ 1 s. We observed significant length contraction: As t was decreased, participants perceived the two distances as equal only when the comparison distance was made progressively greater than the reference distance. Furthermore, the use of weaker taps significantly enhanced participants' length contraction. These findings confirm the model's predictions, supporting the view that the spatiotemporal percept is a best estimate resulting from a Bayesian inference process. PMID:27121574
Bayesian inference on genetic merit under uncertain paternity
Tempelman Robert J
2003-09-01
Full Text Available Abstract A hierarchical animal model was developed for inference on genetic merit of livestock with uncertain paternity. Fully conditional posterior distributions for fixed and genetic effects, variance components, sire assignments and their probabilities are derived to facilitate a Bayesian inference strategy using MCMC methods. We compared this model to a model based on the Henderson average numerator relationship (ANRM in a simulation study with 10 replicated datasets generated for each of two traits. Trait 1 had a medium heritability (h2 for each of direct and maternal genetic effects whereas Trait 2 had a high h2 attributable only to direct effects. The average posterior probabilities inferred on the true sire were between 1 and 10% larger than the corresponding priors (the inverse of the number of candidate sires in a mating pasture for Trait 1 and between 4 and 13% larger than the corresponding priors for Trait 2. The predicted additive and maternal genetic effects were very similar using both models; however, model choice criteria (Pseudo Bayes Factor and Deviance Information Criterion decisively favored the proposed hierarchical model over the ANRM model.
Type Ia Supernova Light Curve Inference: Hierarchical Bayesian Analysis in the Near Infrared
Mandel, Kaisey S; Friedman, Andrew S; Kirshner, Robert P
2009-01-01
We present a comprehensive statistical analysis of the properties of Type Ia SN light curves in the near infrared using recent data from PAIRITEL and the literature. We construct a hierarchical Bayesian framework, incorporating several uncertainties including photometric error, peculiar velocities, dust extinction and intrinsic variations, for coherent statistical inference. SN Ia light curve inferences are drawn from the global posterior probability of parameters describing both individual supernovae and the population conditioned on the entire SN Ia NIR dataset. The logical structure of the hierarchical Bayesian model is represented by a directed acyclic graph. Fully Bayesian analysis of the model and data is enabled by an efficient MCMC algorithm exploiting the conditional structure using Gibbs sampling. We apply this framework to the JHK_s SN Ia light curve data. A new light curve model captures the observed J-band light curve shape variations. The intrinsic variances in peak absolute magnitudes are: sigm...
BAMBI: blind accelerated multimodal Bayesian inference
Graff, Philip; Hobson, Michael P; Lasenby, Anthony
2011-01-01
In this paper we present an algorithm for rapid Bayesian analysis that combines the benefits of nested sampling and artificial neural networks. The blind accelerated multimodal Bayesian inference (BAMBI) algorithm implements the MultiNest package for nested sampling as well as the training of an artificial neural network (NN) to learn the likelihood function. In the case of computationally expensive likelihoods, this allows the substitution of a much more rapid approximation in order to increase significantly the speed of the analysis. We begin by demonstrating, with a few toy examples, the ability of a NN to learn complicated likelihood surfaces. BAMBI's ability to decrease running time for Bayesian inference is then demonstrated in the context of estimating cosmological parameters from WMAP and other observations. We show that valuable speed increases are achieved in addition to obtaining NNs trained on the likelihood functions for the different model and data combinations. These NNs can then be used for an...
Bayesianism and inference to the best explanation
Valeriano IRANZO
2008-01-01
Full Text Available Bayesianism and Inference to the best explanation (IBE are two different models of inference. Recently there has been some debate about the possibility of “bayesianizing” IBE. Firstly I explore several alternatives to include explanatory considerations in Bayes’s Theorem. Then I distinguish two different interpretations of prior probabilities: “IBE-Bayesianism” (IBE-Bay and “frequentist-Bayesianism” (Freq-Bay. After detailing the content of the latter, I propose a rule for assessing the priors. I also argue that Freq-Bay: (i endorses a role for explanatory value in the assessment of scientific hypotheses; (ii avoids a purely subjectivist reading of prior probabilities; and (iii fits better than IBE-Bayesianism with two basic facts about science, i.e., the prominent role played by empirical testing and the existence of many scientific theories in the past that failed to fulfil their promises and were subsequently abandoned.
The NIFTY way of Bayesian signal inference
We introduce NIFTY, 'Numerical Information Field Theory', a software package for the development of Bayesian signal inference algorithms that operate independently from any underlying spatial grid and its resolution. A large number of Bayesian and Maximum Entropy methods for 1D signal reconstruction, 2D imaging, as well as 3D tomography, appear formally similar, but one often finds individualized implementations that are neither flexible nor easily transferable. Signal inference in the framework of NIFTY can be done in an abstract way, such that algorithms, prototyped in 1D, can be applied to real world problems in higher-dimensional settings. NIFTY as a versatile library is applicable and already has been applied in 1D, 2D, 3D and spherical settings. A recent application is the D3PO algorithm targeting the non-trivial task of denoising, deconvolving, and decomposing photon observations in high energy astronomy
Using Alien Coins to Test Whether Simple Inference Is Bayesian
Cassey, Peter; Hawkins, Guy E.; Donkin, Chris; Brown, Scott D.
2016-01-01
Reasoning and inference are well-studied aspects of basic cognition that have been explained as statistically optimal Bayesian inference. Using a simplified experimental design, we conducted quantitative comparisons between Bayesian inference and human inference at the level of individuals. In 3 experiments, with more than 13,000 participants, we…
Constrained bayesian inference of project performance models
Sunmola, Funlade
2013-01-01
Project performance models play an important role in the management of project success. When used for monitoring projects, they can offer predictive ability such as indications of possible delivery problems. Approaches for monitoring project performance relies on available project information including restrictions imposed on the project, particularly the constraints of cost, quality, scope and time. We study in this paper a Bayesian inference methodology for project performance modelling in ...
Bayesian approach to change point detection of unemployment rate via MCMC methods
Reisnerová, Soňa
Plzeň : University of West Bohemia in Pilsen, 2006 - (Lukáš, L.), s. 447-452 ISBN 978-80-7043-480-2. [Mathematical Methods in Economics 2006. Plzeň (CZ), 13.09.2006-15.09.2006] R&D Projects: GA AV ČR IAA101120604 Institutional research plan: CEZ:AV0Z10750506 Keywords : Change point * unemployment rate * MCMC * poisson model Subject RIV: BB - Applied Statistics, Operational Research
Koepke, C.; Irving, J.
2015-12-01
Bayesian solutions to inverse problems in near-surface geophysics and hydrology have gained increasing popularity as a means of estimating not only subsurface model parameters, but also their corresponding uncertainties that can be used in probabilistic forecasting and risk analysis. In particular, Markov-chain-Monte-Carlo (MCMC) methods have attracted much recent attention as a means of statistically sampling from the Bayesian posterior distribution. In this regard, two approaches are commonly used to improve the computational tractability of the Bayesian-MCMC approach: (i) Forward models involving a simplification of the underlying physics are employed, which offer a significant reduction in the time required to calculate data, but generally at the expense of model accuracy, and (ii) the model parameter space is represented using a limited set of spatially correlated basis functions as opposed to a more intuitive high-dimensional pixel-based parameterization. It has become well understood that model inaccuracies resulting from (i) can lead to posterior parameter distributions that are highly biased and overly confident. Further, when performing model reduction as described in (ii), it is not clear how the prior distribution for the basis weights should be defined because simple (e.g., Gaussian or uniform) priors that may be suitable for a pixel-based parameterization may result in a strong prior bias when used for the weights. To address the issue of model error resulting from known forward model approximations, we generate a set of error training realizations and analyze them with principal component analysis (PCA) in order to generate a sparse basis. The latter is used in the MCMC inversion to remove the main model-error component from the residuals. To improve issues related to prior bias when performing model reduction, we also use a training realization approach, but this time models are simulated from the prior distribution and analyzed using independent
Murakami, Yohei; Takada, Shoji
2013-01-01
When exact values of model parameters in systems biology are not available from experiments, they need to be inferred so that the resulting simulation reproduces the experimentally known phenomena. For the purpose, Bayesian statistics with Markov chain Monte Carlo (MCMC) is a useful method. Biological experiments are often performed with cell population, and the results are represented by histograms. On another front, experiments sometimes indicate the existence of a specific bifurcation patt...
Bayesian Estimation and Inference Using Stochastic Electronics.
Thakur, Chetan Singh; Afshar, Saeed; Wang, Runchun M; Hamilton, Tara J; Tapson, Jonathan; van Schaik, André
2016-01-01
In this paper, we present the implementation of two types of Bayesian inference problems to demonstrate the potential of building probabilistic algorithms in hardware using single set of building blocks with the ability to perform these computations in real time. The first implementation, referred to as the BEAST (Bayesian Estimation and Stochastic Tracker), demonstrates a simple problem where an observer uses an underlying Hidden Markov Model (HMM) to track a target in one dimension. In this implementation, sensors make noisy observations of the target position at discrete time steps. The tracker learns the transition model for target movement, and the observation model for the noisy sensors, and uses these to estimate the target position by solving the Bayesian recursive equation online. We show the tracking performance of the system and demonstrate how it can learn the observation model, the transition model, and the external distractor (noise) probability interfering with the observations. In the second implementation, referred to as the Bayesian INference in DAG (BIND), we show how inference can be performed in a Directed Acyclic Graph (DAG) using stochastic circuits. We show how these building blocks can be easily implemented using simple digital logic gates. An advantage of the stochastic electronic implementation is that it is robust to certain types of noise, which may become an issue in integrated circuit (IC) technology with feature sizes in the order of tens of nanometers due to their low noise margin, the effect of high-energy cosmic rays and the low supply voltage. In our framework, the flipping of random individual bits would not affect the system performance because information is encoded in a bit stream. PMID:27047326
Trans-Dimensional Bayesian Inference for Gravitational Lens Substructures
Brewer, Brendon J; Lewis, Geraint F
2015-01-01
We introduce a Bayesian solution to the problem of inferring the density profile of strong gravitational lenses when the lens galaxy may contain multiple dark or faint substructures. The source and lens models are based on a superposition of an unknown number of non-negative basis functions (or "blobs") whose form was chosen with speed as a primary criterion. The prior distribution for the blobs' properties is specified hierarchically, so the mass function of substructures is a natural output of the method. We use reversible jump Markov Chain Monte Carlo (MCMC) within Diffusive Nested Sampling (DNS) to sample the posterior distribution and evaluate the marginal likelihood of the model, including the summation over the unknown number of blobs in the source and the lens. We demonstrate the method on a simulated data set with a single substructure, which is recovered well with moderate uncertainties. We also apply the method to the g-band image of the "Cosmic Horseshoe" system, and find some hints of potential s...
Bayesian Posterior Distributions Without Markov Chains
Cole, Stephen R.; Chu, Haitao; Greenland, Sander; Hamra, Ghassan; Richardson, David B.
2012-01-01
Bayesian posterior parameter distributions are often simulated using Markov chain Monte Carlo (MCMC) methods. However, MCMC methods are not always necessary and do not help the uninitiated understand Bayesian inference. As a bridge to understanding Bayesian inference, the authors illustrate a transparent rejection sampling method. In example 1, they illustrate rejection sampling using 36 cases and 198 controls from a case-control study (1976–1983) assessing the relation between residential ex...
Human collective intelligence as distributed Bayesian inference
Krafft, Peter M; Pan, Wei; Della Penna, Nicolás; Altshuler, Yaniv; Shmueli, Erez; Tenenbaum, Joshua B; Pentland, Alex
2016-01-01
Collective intelligence is believed to underly the remarkable success of human society. The formation of accurate shared beliefs is one of the key components of human collective intelligence. How are accurate shared beliefs formed in groups of fallible individuals? Answering this question requires a multiscale analysis. We must understand both the individual decision mechanisms people use, and the properties and dynamics of those mechanisms in the aggregate. As of yet, mathematical tools for such an approach have been lacking. To address this gap, we introduce a new analytical framework: We propose that groups arrive at accurate shared beliefs via distributed Bayesian inference. Distributed inference occurs through information processing at the individual level, and yields rational belief formation at the group level. We instantiate this framework in a new model of human social decision-making, which we validate using a dataset we collected of over 50,000 users of an online social trading platform where inves...
Bayesian Inference of a Multivariate Regression Model
Marick S. Sinay
2014-01-01
Full Text Available We explore Bayesian inference of a multivariate linear regression model with use of a flexible prior for the covariance structure. The commonly adopted Bayesian setup involves the conjugate prior, multivariate normal distribution for the regression coefficients and inverse Wishart specification for the covariance matrix. Here we depart from this approach and propose a novel Bayesian estimator for the covariance. A multivariate normal prior for the unique elements of the matrix logarithm of the covariance matrix is considered. Such structure allows for a richer class of prior distributions for the covariance, with respect to strength of beliefs in prior location hyperparameters, as well as the added ability, to model potential correlation amongst the covariance structure. The posterior moments of all relevant parameters of interest are calculated based upon numerical results via a Markov chain Monte Carlo procedure. The Metropolis-Hastings-within-Gibbs algorithm is invoked to account for the construction of a proposal density that closely matches the shape of the target posterior distribution. As an application of the proposed technique, we investigate a multiple regression based upon the 1980 High School and Beyond Survey.
Universal Darwinism as a process of Bayesian inference
Campbell, John O
2016-01-01
Many of the mathematical frameworks describing natural selection are equivalent to Bayes Theorem, also known as Bayesian updating. By definition, a process of Bayesian Inference is one which involves a Bayesian update, so we may conclude that these frameworks describe natural selection as a process of Bayesian inference. Thus natural selection serves as a counter example to a widely-held interpretation that restricts Bayesian Inference to human mental processes (including the endeavors of statisticians). As Bayesian inference can always be cast in terms of (variational) free energy minimization, natural selection can be viewed as comprising two components: a generative model of an "experiment" in the external world environment, and the results of that "experiment" or the "surprise" entailed by predicted and actual outcomes of the "experiment". Minimization of free energy implies that the implicit measure of "surprise" experienced serves to update the generative model in a Bayesian manner. This description clo...
Efficient Bayesian inference for ARFIMA processes
Graves, T.; Gramacy, R. B.; Franzke, C. L. E.; Watkins, N. W.
2015-03-01
Many geophysical quantities, like atmospheric temperature, water levels in rivers, and wind speeds, have shown evidence of long-range dependence (LRD). LRD means that these quantities experience non-trivial temporal memory, which potentially enhances their predictability, but also hampers the detection of externally forced trends. Thus, it is important to reliably identify whether or not a system exhibits LRD. In this paper we present a modern and systematic approach to the inference of LRD. Rather than Mandelbrot's fractional Gaussian noise, we use the more flexible Autoregressive Fractional Integrated Moving Average (ARFIMA) model which is widely used in time series analysis, and of increasing interest in climate science. Unlike most previous work on the inference of LRD, which is frequentist in nature, we provide a systematic treatment of Bayesian inference. In particular, we provide a new approximate likelihood for efficient parameter inference, and show how nuisance parameters (e.g. short memory effects) can be integrated over in order to focus on long memory parameters, and hypothesis testing more directly. We illustrate our new methodology on the Nile water level data, with favorable comparison to the standard estimators.
A Fast Iterative Bayesian Inference Algorithm for Sparse Channel Estimation
Pedersen, Niels Lovmand; Manchón, Carles Navarro; Fleury, Bernard Henri
2013-01-01
representation of the Bessel K probability density function; a highly efficient, fast iterative Bayesian inference method is then applied to the proposed model. The resulting estimator outperforms other state-of-the-art Bayesian and non-Bayesian estimators, either by yielding lower mean squared estimation error...
Optical tweezers calibration with Bayesian inference
Türkcan, Silvan; Richly, Maximilian U.; Le Gall, Antoine; Fiszman, Nicolas; Masson, Jean-Baptiste; Westbrook, Nathalie; Perronet, Karen; Alexandrou, Antigoni
2014-09-01
We present a new method for calibrating an optical-tweezer setup that is based on Bayesian inference1. This method employs an algorithm previously used to analyze the confined trajectories of receptors within lipid rafts2,3. The main advantages of this method are that it does not require input parameters and is insensitive to systematic errors like the drift of the setup. Additionally, it exploits a much larger amount of the information stored in the recorded bead trajectory than standard calibration approaches. The additional information can be used to detect deviations from the perfect harmonic potential or detect environmental influences on the bead. The algorithm infers the diffusion coefficient and the potential felt by a trapped bead, and only requires the bead trajectory as input. We demonstrate that this method outperforms the equipartition method and the power-spectrum method in input information required (bead radius and trajectory length) and in output accuracy. Furthermore, by inferring a higher order potential our method can reveal deviations from the assumed second-order potential. More generally, this method can also be used for magnetic-tweezer calibration.
Strategies for MCMC computation inquantitative genetics
Waagepetersen, Rasmus; Ibánēz-Escriche, Noelia; Sorensen, Daniel
likelihood inference is complicated since it is not possible to evaluate explicitly the likelihood function and conventional Gibbs sampling is difficult since the full conditional distributions are not anymore of standard forms. The aim of this paper is to discuss strategies to obtain efficient Markov chain...... sparse inverse. Maximum likelihood inference and Bayesian inference for the linear mixed model are well-studied topics (Sorensen and Gianola, 2002). Regarding Bayesian inference, with appropriate choice of priors, the full conditional distributions are standard distributions and Gibbs sampling can be...... both in size and regarding the inferences concerning the genetic covariance parameters. Section 2 discusses general strategies for obtaining efficient MCMC algorithms while Section 3 considers these strategies in the specific context of the San Cristobal-Gaudy et al. (1998) model. Section 4 presents...
Bayesian electron density inference from JET lithium beam emission spectra using Gaussian processes
Kwak, Sehyun; Brix, M; Ghim, Y -c
2016-01-01
A Bayesian model to infer edge electron density profiles is developed for the JET lithium beam emission spectroscopy system, measuring Li I line radiation using 26 channels with ~1 cm spatial resolution and 10~20 ms temporal resolution. The density profile is modelled using a Gaussian process prior, and the uncertainty of the density profile is calculated by a Markov Chain Monte Carlo (MCMC) scheme. From the spectra measured by the transmission grating spectrometer, the Li line intensities are extracted, and modelled as a function of the plasma density by a multi-state model which describes the relevant processes between neutral lithium beam atoms and plasma particles. The spectral model fully takes into account interference filter and instrument effects, that are separately estimated, again using Gaussian processes. The line intensities are inferred based on a spectral model consistent with the measured spectra within their uncertainties, which includes photon statistics and electronic noise. Our newly devel...
Universal Darwinism As a Process of Bayesian Inference.
Campbell, John O
2016-01-01
Many of the mathematical frameworks describing natural selection are equivalent to Bayes' Theorem, also known as Bayesian updating. By definition, a process of Bayesian Inference is one which involves a Bayesian update, so we may conclude that these frameworks describe natural selection as a process of Bayesian inference. Thus, natural selection serves as a counter example to a widely-held interpretation that restricts Bayesian Inference to human mental processes (including the endeavors of statisticians). As Bayesian inference can always be cast in terms of (variational) free energy minimization, natural selection can be viewed as comprising two components: a generative model of an "experiment" in the external world environment, and the results of that "experiment" or the "surprise" entailed by predicted and actual outcomes of the "experiment." Minimization of free energy implies that the implicit measure of "surprise" experienced serves to update the generative model in a Bayesian manner. This description closely accords with the mechanisms of generalized Darwinian process proposed both by Dawkins, in terms of replicators and vehicles, and Campbell, in terms of inferential systems. Bayesian inference is an algorithm for the accumulation of evidence-based knowledge. This algorithm is now seen to operate over a wide range of evolutionary processes, including natural selection, the evolution of mental models and cultural evolutionary processes, notably including science itself. The variational principle of free energy minimization may thus serve as a unifying mathematical framework for universal Darwinism, the study of evolutionary processes operating throughout nature. PMID:27375438
Mira Antonietta; Tenconi Paolo; Bressanini Dario
2003-01-01
We propose a general purpose variance reduction technique for MCMC estimators. The idea is obtained by combining standard variance reduction principles known for regular Monte Carlo simulations (Ripley, 1987) and the Zero-Variance principle introduced in the physics literature (Assaraf and Caffarel, 1999). The potential of the new idea is illustrated with some toy examples and an application to Bayesian estimation
De Freitas, Nando; Hojen-Sorensen, Pedro; Jordan, Michael I.; Russell, Stuart
2013-01-01
We propose a new class of learning algorithms that combines variational approximation and Markov chain Monte Carlo (MCMC) simulation. Naive algorithms that use the variational approximation as proposal distribution can perform poorly because this approximation tends to underestimate the true variance and other features of the data. We solve this problem by introducing more sophisticated MCMC algorithms. One of these algorithms is a mixture of two MCMC kernels: a random walk Metropolis kernel ...
A new method for the experimental determination of stopping powers based on Bayesian Inference with the Markov chain Monte Carlo (MCMC) algorithm has been devised. This method avoids the difficulties related to thin target preparation. By measuring the RBS spectra for a known material, and using the known underlying physics, the stopping powers are determined by best matching the simulated spectra with the experimental spectra. Using silicon, SiO2 and Al2O3 as test cases, good agreement is obtained between calculated and experimental data. (author)
Sparsity has become a key concept for solving of high-dimensional inverse problems using variational regularization techniques. Recently, using similar sparsity-constraints in the Bayesian framework for inverse problems by encoding them in the prior distribution has attracted attention. Important questions about the relation between regularization theory and Bayesian inference still need to be addressed when using sparsity promoting inversion. A practical obstacle for these examinations is the lack of fast posterior sampling algorithms for sparse, high-dimensional Bayesian inversion. Accessing the full range of Bayesian inference methods requires being able to draw samples from the posterior probability distribution in a fast and efficient way. This is usually done using Markov chain Monte Carlo (MCMC) sampling algorithms. In this paper, we develop and examine a new implementation of a single component Gibbs MCMC sampler for sparse priors relying on L1-norms. We demonstrate that the efficiency of our Gibbs sampler increases when the level of sparsity or the dimension of the unknowns is increased. This property is contrary to the properties of the most commonly applied Metropolis–Hastings (MH) sampling schemes. We demonstrate that the efficiency of MH schemes for L1-type priors dramatically decreases when the level of sparsity or the dimension of the unknowns is increased. Practically, Bayesian inversion for L1-type priors using MH samplers is not feasible at all. As this is commonly believed to be an intrinsic feature of MCMC sampling, the performance of our Gibbs sampler also challenges common beliefs about the applicability of sample based Bayesian inference. (paper)
Bayesian Inference and Optimal Design in the Sparse Linear Model
Seeger, Matthias; Steinke, Florian; Tsuda, Koji
2007-01-01
The sparse linear model has seen many successful applications in Statistics, Machine Learning, and Computational Biology, such as identification of gene regulatory networks from micro-array expression data. Prior work has either approximated Bayesian inference by expensive Markov chain Monte Carlo, or replaced it by point estimation. We show how to obtain a good approximation to Bayesian analysis efficiently, using the Expectation Propagation method. We also address the problems of optimal de...
Kernel Approximate Bayesian Computation for Population Genetic Inferences
Nakagome, Shigeki; Fukumizu, Kenji; Mano, Shuhei
2012-01-01
Approximate Bayesian computation (ABC) is a likelihood-free approach for Bayesian inferences based on a rejection algorithm method that applies a tolerance of dissimilarity between summary statistics from observed and simulated data. Although several improvements to the algorithm have been proposed, none of these improvements avoid the following two sources of approximation: 1) lack of sufficient statistics: sampling is not from the true posterior density given data but from an approximate po...
Methods for Bayesian power spectrum inference with galaxy surveys
Jasche, Jens
2013-01-01
We derive and implement a full Bayesian large scale structure inference method aiming at precision recovery of the cosmological power spectrum from galaxy redshift surveys. Our approach improves over previous Bayesian methods by performing a joint inference of the three dimensional density field, the cosmological power spectrum, luminosity dependent galaxy biases and corresponding normalizations. We account for all joint and correlated uncertainties between all inferred quantities. Classes of galaxies with different biases are treated as separate sub samples. The method therefore also allows the combined analysis of more than one galaxy survey. In particular, it solves the problem of inferring the power spectrum from galaxy surveys with non-trivial survey geometries by exploring the joint posterior distribution with efficient implementations of multiple block Markov chain and Hybrid Monte Carlo methods. Our Markov sampler achieves high statistical efficiency in low signal to noise regimes by using a determini...
Approximate Bayesian inference for complex ecosystems
Michael P H Stumpf
2014-01-01
Mathematical models have been central to ecology for nearly a century. Simple models of population dynamics have allowed us to understand fundamental aspects underlying the dynamics and stability of ecological systems. What has remained a challenge, however, is to meaningfully interpret experimental or observational data in light of mathematical models. Here, we review recent developments, notably in the growing field of approximate Bayesian computation (ABC), that allow us to calibrate mathe...
A Bayesian Approach to Protein Inference Problem in Shotgun Proteomics
Li, Yong Fuga; Arnold, Randy J.; Li, Yixue; Radivojac, Predrag; Sheng, Quanhu; Tang, Haixu
2009-01-01
The protein inference problem represents a major challenge in shotgun proteomics. In this article, we describe a novel Bayesian approach to address this challenge by incorporating the predicted peptide detectabilities as the prior probabilities of peptide identification. We propose a rigorious probabilistic model for protein inference and provide practical algoritmic solutions to this problem. We used a complex synthetic protein mixture to test our method and obtained promising results.
Bayesian Inference of Kinematics and Memberships of Open Cluster
Shao, Z. Y.; Chen, L.; Zhong, J.; Hou, J. L.
2014-07-01
Based on the Bayesian Inference (BI) method, the Multiple-modelling approach is improved to combine coordinative positions, proper motions (PM) and radial velocities (RV), to separate the motion of the open cluster from field stars, as well as to describe the intrinsic kinematic status of the cluster.
Pig Data and Bayesian Inference on Multinomial Probabilities
Kern, John C.
2006-01-01
Bayesian inference on multinomial probabilities is conducted based on data collected from the game Pass the Pigs[R]. Prior information on these probabilities is readily available from the instruction manual, and is easily incorporated in a Dirichlet prior. Posterior analysis of the scoring probabilities quantifies the discrepancy between empirical…
Stan: A Probabilistic Programming Language for Bayesian Inference and Optimization
Gelman, Andrew; Lee, Daniel; Guo, Jiqiang
2015-01-01
Stan is a free and open-source C++ program that performs Bayesian inference or optimization for arbitrary user-specified models and can be called from the command line, R, Python, Matlab, or Julia and has great promise for fitting large and complex statistical models in many areas of application. We discuss Stan from users' and developers'…
Bayesian inference model for fatigue life of laminated composites
Dimitrov, Nikolay Krasimirov; Kiureghian, Armen Der; Berggreen, Christian
2016-01-01
A probabilistic model for estimating the fatigue life of laminated composite plates is developed. The model is based on lamina-level input data, making it possible to predict fatigue properties for a wide range of laminate configurations. Model parameters are estimated by Bayesian inference. The...
Bayesian inference of chemical kinetic models from proposed reactions
Galagali, Nikhil
2015-02-01
© 2014 Elsevier Ltd. Bayesian inference provides a natural framework for combining experimental data with prior knowledge to develop chemical kinetic models and quantify the associated uncertainties, not only in parameter values but also in model structure. Most existing applications of Bayesian model selection methods to chemical kinetics have been limited to comparisons among a small set of models, however. The significant computational cost of evaluating posterior model probabilities renders traditional Bayesian methods infeasible when the model space becomes large. We present a new framework for tractable Bayesian model inference and uncertainty quantification using a large number of systematically generated model hypotheses. The approach involves imposing point-mass mixture priors over rate constants and exploring the resulting posterior distribution using an adaptive Markov chain Monte Carlo method. The posterior samples are used to identify plausible models, to quantify rate constant uncertainties, and to extract key diagnostic information about model structure-such as the reactions and operating pathways most strongly supported by the data. We provide numerical demonstrations of the proposed framework by inferring kinetic models for catalytic steam and dry reforming of methane using available experimental data.
Bayesian Inference in the Modern Design of Experiments
DeLoach, Richard
2008-01-01
This paper provides an elementary tutorial overview of Bayesian inference and its potential for application in aerospace experimentation in general and wind tunnel testing in particular. Bayes Theorem is reviewed and examples are provided to illustrate how it can be applied to objectively revise prior knowledge by incorporating insights subsequently obtained from additional observations, resulting in new (posterior) knowledge that combines information from both sources. A logical merger of Bayesian methods and certain aspects of Response Surface Modeling is explored. Specific applications to wind tunnel testing, computational code validation, and instrumentation calibration are discussed.
Zhang, Guannan [ORNL; Webster, Clayton G [ORNL; Gunzburger, Max D [ORNL
2012-09-01
Although Bayesian analysis has become vital to the quantification of prediction uncertainty in groundwater modeling, its application has been hindered due to the computational cost associated with numerous model executions needed for exploring the posterior probability density function (PPDF) of model parameters. This is particularly the case when the PPDF is estimated using Markov Chain Monte Carlo (MCMC) sampling. In this study, we develop a new approach that improves computational efficiency of Bayesian inference by constructing a surrogate system based on an adaptive sparse-grid high-order stochastic collocation (aSG-hSC) method. Unlike previous works using first-order hierarchical basis, we utilize a compactly supported higher-order hierar- chical basis to construct the surrogate system, resulting in a significant reduction in the number of computational simulations required. In addition, we use hierarchical surplus as an error indi- cator to determine adaptive sparse grids. This allows local refinement in the uncertain domain and/or anisotropic detection with respect to the random model parameters, which further improves computational efficiency. Finally, we incorporate a global optimization technique and propose an iterative algorithm for building the surrogate system for the PPDF with multiple significant modes. Once the surrogate system is determined, the PPDF can be evaluated by sampling the surrogate system directly with very little computational cost. The developed method is evaluated first using a simple analytical density function with multiple modes and then using two synthetic groundwater reactive transport models. The groundwater models represent different levels of complexity; the first example involves coupled linear reactions and the second example simulates nonlinear ura- nium surface complexation. The results show that the aSG-hSC is an effective and efficient tool for Bayesian inference in groundwater modeling in comparison with conventional
An equilibrium validation technique based on Bayesian inference
In recent years, Bayesian probability theory has been used in a number of experiments to fold uncertainties and interdependences in the diagnostic data and forward models, and together with prior knowledge of the state of the plasma, thus increase accuracy of inferred physics variables. Key developments include the application to current and flux surface tomography, effective charge, the electron energy distribution function, neutron spectrometry and density. Virtual observations have also been introduced to better constrain inferred quantities in current tomography. In this work we present Bayesian inference results of toroidal and poloidal current and flux surface tomography. Whilst the uncertainty in these profiles, as well as the uncertainty in inferred parameters such as the safety factor profile is small (<5%), the inference can change substantially depending on the physics model used. We also present Bayesian inference results for Thomson scattering and charge-exchange recombination spectroscopy. In separate work we have computed radial force balance components on the midplane in the Mega Ampere Spherical tokamak. Our aim is to establish a validation framework for different equilibrium physics models. We find that in the overlapping region of the core (normalized poloidal flux less than 0.4) and motional Stark effect (MSE) chords, the plasma is consistent with static Grad-Shafranov force balance to within two standard deviations. In the outboard edge region, where MSE data are also available, the pressure gradient exceeds the Lorentz force. Most likely, this is because the poloidal current is not constrained to zero at the plasma edge. To lowest order, the results suggest computing components of force balance are useful to assess data-consistency, independent of any equilibrium solution. To first order, we have integrated the residue to force balance to infer an energetic particle pressure.
Dimension-independent likelihood-informed MCMC
Cui, Tiangang
2015-10-08
Many Bayesian inference problems require exploring the posterior distribution of high-dimensional parameters that represent the discretization of an underlying function. This work introduces a family of Markov chain Monte Carlo (MCMC) samplers that can adapt to the particular structure of a posterior distribution over functions. Two distinct lines of research intersect in the methods developed here. First, we introduce a general class of operator-weighted proposal distributions that are well defined on function space, such that the performance of the resulting MCMC samplers is independent of the discretization of the function. Second, by exploiting local Hessian information and any associated low-dimensional structure in the change from prior to posterior distributions, we develop an inhomogeneous discretization scheme for the Langevin stochastic differential equation that yields operator-weighted proposals adapted to the non-Gaussian structure of the posterior. The resulting dimension-independent and likelihood-informed (DILI) MCMC samplers may be useful for a large class of high-dimensional problems where the target probability measure has a density with respect to a Gaussian reference measure. Two nonlinear inverse problems are used to demonstrate the efficiency of these DILI samplers: an elliptic PDE coefficient inverse problem and path reconstruction in a conditioned diffusion.
Höhna, Sebastian; Landis, Michael J; Heath, Tracy A; Boussau, Bastien; Lartillot, Nicolas; Moore, Brian R; Huelsenbeck, John P; Ronquist, Fredrik
2016-07-01
Programs for Bayesian inference of phylogeny currently implement a unique and ﬁxed suite of models. Consequently, users of these software packages are simultaneously forced to use a number of programs for a given study, while also lacking the freedom to explore models that have not been implemented by the developers of those programs. We developed a new open-source software package, RevBayes, to address these problems. RevBayes is entirely based on probabilistic graphical models, a powerful generic framework for specifying and analyzing statistical models. Phylogenetic-graphical models can be speciﬁed interactively in RevBayes, piece by piece, using a new succinct and intuitive language called Rev. Rev is similar to the R language and the BUGS model-speciﬁcation language, and should be easy to learn for most users. The strength of RevBayes is the simplicity with which one can design, specify, and implement new and complex models. Fortunately, this tremendous ﬂexibility does not come at the cost of slower computation; as we demonstrate, RevBayes outperforms competing software for several standard analyses. Compared with other programs, RevBayes has fewer black-box elements. Users need to explicitly specify each part of the model and analysis. Although this explicitness may initially be unfamiliar, we are convinced that this transparency will improve understanding of phylogenetic models in our ﬁeld. Moreover, it will motivate the search for improvements to existing methods by brazenly exposing the model choices that we make to critical scrutiny. RevBayes is freely available at http://www.RevBayes.com [Bayesian inference; Graphical models; MCMC; statistical phylogenetics.]. PMID:27235697
Revealing ecological networks using Bayesian network inference algorithms.
Milns, Isobel; Beale, Colin M; Smith, V Anne
2010-07-01
Understanding functional relationships within ecological networks can help reveal keys to ecosystem stability or fragility. Revealing these relationships is complicated by the difficulties of isolating variables or performing experimental manipulations within a natural ecosystem, and thus inferences are often made by matching models to observational data. Such models, however, require assumptions-or detailed measurements-of parameters such as birth and death rate, encounter frequency, territorial exclusion, and predation success. Here, we evaluate the use of a Bayesian network inference algorithm, which can reveal ecological networks based upon species and habitat abundance alone. We test the algorithm's performance and applicability on observational data of avian communities and habitat in the Peak District National Park, United Kingdom. The resulting networks correctly reveal known relationships among habitat types and known interspecific relationships. In addition, the networks produced novel insights into ecosystem structure and identified key species with high connectivity. Thus, Bayesian networks show potential for becoming a valuable tool in ecosystem analysis. PMID:20715607
Parsing optical scanned 3D data by Bayesian inference
Xiong, Hanwei; Xu, Jun; Xu, Chenxi; Pan, Ming
2015-10-01
Optical devices are always used to digitize complex objects to get their shapes in form of point clouds. The results have no semantic meaning about the objects, and tedious process is indispensable to segment the scanned data to get meanings. The reason for a person to perceive an object correctly is the usage of knowledge, so Bayesian inference is used to the goal. A probabilistic And-Or-Graph is used as a unified framework of representation, learning, and recognition for a large number of object categories, and a probabilistic model defined on this And-Or-Graph is learned from a relatively small training set per category. Given a set of 3D scanned data, the Bayesian inference constructs a most probable interpretation of the object, and a semantic segment is obtained from the part decomposition. Some examples are given to explain the method.
Fast Bayesian inference for slow-roll inflation
Ringeval, Christophe
2013-01-01
We present and discuss a new approach increasing by orders of magnitude the speed of performing Bayesian inference and parameter estimation within the framework of slow-roll inflation. The method relies on the determination of an effective likelihood for inflation which is a function of the primordial amplitude of the scalar perturbations complemented with the necessary number of the so-called Hubble flow functions to reach the desired accuracy. Starting from any cosmological data set, the effective likelihood is obtained by marginalisation over the standard cosmological parameters, here viewed as "nuisance" from the early Universe point of view. As being low-dimensional, basic machine-learning algorithms can be trained to accurately reproduce its multidimensional shape and then be used as a proxy to perform fast Bayesian inference on the inflationary models. The robustness and accuracy of the method are illustrated using the Planck Cosmic Microwave Background (CMB) data to perform primordial parameter estima...
Bayesian Inference for Smoking Cessation with a Latent Cure State
Luo, Sheng; Crainiceanu, Ciprian M.; Thomas A Louis; Chatterjee, Nilanjan
2009-01-01
We present a Bayesian approach to modeling dynamic smoking addiction behavior processes when cure is not directly observed due to censoring. Subject-specific probabilities model the stochastic transitions among three behavioral states: smoking, transient quitting, and permanent quitting (absorbent state). A multivariate normal distribution for random effects is used to account for the potential correlation among the subject-specific transition probabilities. Inference is conducted using a Bay...
Practical Statistics for LHC Physicists: Bayesian Inference (3/3)
CERN. Geneva
2015-01-01
These lectures cover those principles and practices of statistics that are most relevant for work at the LHC. The first lecture discusses the basic ideas of descriptive statistics, probability and likelihood. The second lecture covers the key ideas in the frequentist approach, including confidence limits, profile likelihoods, p-values, and hypothesis testing. The third lecture covers inference in the Bayesian approach. Throughout, real-world examples will be used to illustrate the practical application of the ideas. No previous knowledge is assumed.
Bayesian Fusion Algorithm for Inferring Trust in Wireless Sensor Networks
Mohammad Momani; Subhash Challa; Rami Alhmouz
2010-01-01
This paper introduces a new Bayesian fusion algorithm to combine more than one trust component (data trust and communication trust) to infer the overall trust between nodes. This research work proposes that one trust component is not enough when deciding on whether or not to trust a specific node in a wireless sensor network. This paper discusses and analyses the results from the communication trust component (binary) and the data trust component (continuous) and proves that either component ...
Halo detection via large-scale Bayesian inference
Merson, Alexander I.; Jasche, Jens; Abdalla, Filipe B.; Lahav, Ofer; Wandelt, Benjamin; Jones, D. Heath; Colless, Matthew
2016-08-01
We present a proof-of-concept of a novel and fully Bayesian methodology designed to detect haloes of different masses in cosmological observations subject to noise and systematic uncertainties. Our methodology combines the previously published Bayesian large-scale structure inference algorithm, HAmiltonian Density Estimation and Sampling algorithm (HADES), and a Bayesian chain rule (the Blackwell-Rao estimator), which we use to connect the inferred density field to the properties of dark matter haloes. To demonstrate the capability of our approach, we construct a realistic galaxy mock catalogue emulating the wide-area 6-degree Field Galaxy Survey, which has a median redshift of approximately 0.05. Application of HADES to the catalogue provides us with accurately inferred three-dimensional density fields and corresponding quantification of uncertainties inherent to any cosmological observation. We then use a cosmological simulation to relate the amplitude of the density field to the probability of detecting a halo with mass above a specified threshold. With this information, we can sum over the HADES density field realisations to construct maps of detection probabilities and demonstrate the validity of this approach within our mock scenario. We find that the probability of successful detection of haloes in the mock catalogue increases as a function of the signal to noise of the local galaxy observations. Our proposed methodology can easily be extended to account for more complex scientific questions and is a promising novel tool to analyse the cosmic large-scale structure in observations.
Methods for Bayesian Power Spectrum Inference with Galaxy Surveys
Jasche, Jens; Wandelt, Benjamin D.
2013-12-01
We derive and implement a full Bayesian large scale structure inference method aiming at precision recovery of the cosmological power spectrum from galaxy redshift surveys. Our approach improves upon previous Bayesian methods by performing a joint inference of the three-dimensional density field, the cosmological power spectrum, luminosity dependent galaxy biases, and corresponding normalizations. We account for all joint and correlated uncertainties between all inferred quantities. Classes of galaxies with different biases are treated as separate subsamples. This method therefore also allows the combined analysis of more than one galaxy survey. In particular, it solves the problem of inferring the power spectrum from galaxy surveys with non-trivial survey geometries by exploring the joint posterior distribution with efficient implementations of multiple block Markov chain and Hybrid Monte Carlo methods. Our Markov sampler achieves high statistical efficiency in low signal-to-noise regimes by using a deterministic reversible jump algorithm. This approach reduces the correlation length of the sampler by several orders of magnitude, turning the otherwise numerically unfeasible problem of joint parameter exploration into a numerically manageable task. We test our method on an artificial mock galaxy survey, emulating characteristic features of the Sloan Digital Sky Survey data release 7, such as its survey geometry and luminosity-dependent biases. These tests demonstrate the numerical feasibility of our large scale Bayesian inference frame work when the parameter space has millions of dimensions. This method reveals and correctly treats the anti-correlation between bias amplitudes and power spectrum, which are not taken into account in current approaches to power spectrum estimation, a 20% effect across large ranges in k space. In addition, this method results in constrained realizations of density fields obtained without assuming the power spectrum or bias parameters
MCMC for non-linear state space models using ensembles of latent sequences
Shestopaloff, Alexander Y.; Neal, Radford M.
2013-01-01
Non-linear state space models are a widely-used class of models for biological, economic, and physical processes. Fitting these models to observed data is a difficult inference problem that has no straightforward solution. We take a Bayesian approach to the inference of unknown parameters of a non-linear state model; this, in turn, requires the availability of efficient Markov Chain Monte Carlo (MCMC) sampling methods for the latent (hidden) variables and model parameters. Using the ensemble ...
Furtado-Junior, I; Abrunhosa, F A; Holanda, F C A F; Tavares, M C S
2016-06-01
Fishing selectivity of the mangrove crab Ucides cordatus in the north coast of Brazil can be defined as the fisherman's ability to capture and select individuals from a certain size or sex (or a combination of these factors) which suggests an empirical selectivity. Considering this hypothesis, we calculated the selectivity curves for males and females crabs using the logit function of the logistic model in the formulation. The Bayesian inference consisted of obtaining the posterior distribution by applying the Markov chain Monte Carlo (MCMC) method to software R using the OpenBUGS, BRugs, and R2WinBUGS libraries. The estimated results of width average carapace selection for males and females compared with previous studies reporting the average width of the carapace of sexual maturity allow us to confirm the hypothesis that most mature individuals do not suffer from fishing pressure; thus, ensuring their sustainability. PMID:26934154
Approximate bayesian parameter inference for dynamical systems in systems biology
This paper proposes to use approximate instead of exact stochastic simulation algorithms for approximate Bayesian parameter inference of dynamical systems in systems biology. It first presents the mathematical framework for the description of systems biology models, especially from the aspect of a stochastic formulation as opposed to deterministic model formulations based on the law of mass action. In contrast to maximum likelihood methods for parameter inference, approximate inference method- share presented which are based on sampling parameters from a known prior probability distribution, which gradually evolves toward a posterior distribution, through the comparison of simulated data from the model to a given data set of measurements. The paper then discusses the simulation process, where an over- view is given of the different exact and approximate methods for stochastic simulation and their improvements that we propose. The exact and approximate simulators are implemented and used within approximate Bayesian parameter inference methods. Our evaluation of these methods on two tasks of parameter estimation in two different models shows that equally good results are obtained much faster when using approximate simulation as compared to using exact simulation. (Author)
A Bayesian Approach to Inferring Rates of Selfing and Locus-Specific Mutation.
Redelings, Benjamin D; Kumagai, Seiji; Tatarenkov, Andrey; Wang, Liuyang; Sakai, Ann K; Weller, Stephen G; Culley, Theresa M; Avise, John C; Uyenoyama, Marcy K
2015-11-01
We present a Bayesian method for characterizing the mating system of populations reproducing through a mixture of self-fertilization and random outcrossing. Our method uses patterns of genetic variation across the genome as a basis for inference about reproduction under pure hermaphroditism, gynodioecy, and a model developed to describe the self-fertilizing killifish Kryptolebias marmoratus. We extend the standard coalescence model to accommodate these mating systems, accounting explicitly for multilocus identity disequilibrium, inbreeding depression, and variation in fertility among mating types. We incorporate the Ewens sampling formula (ESF) under the infinite-alleles model of mutation to obtain a novel expression for the likelihood of mating system parameters. Our Markov chain Monte Carlo (MCMC) algorithm assigns locus-specific mutation rates, drawn from a common mutation rate distribution that is itself estimated from the data using a Dirichlet process prior model. Our sampler is designed to accommodate additional information, including observations pertaining to the sex ratio, the intensity of inbreeding depression, and other aspects of reproduction. It can provide joint posterior distributions for the population-wide proportion of uniparental individuals, locus-specific mutation rates, and the number of generations since the most recent outcrossing event for each sampled individual. Further, estimation of all basic parameters of a given model permits estimation of functions of those parameters, including the proportion of the gene pool contributed by each sex and relative effective numbers. PMID:26374460
Bayesian Top-Down Protein Sequence Alignment with Inferred Position-Specific Gap Penalties.
Neuwald, Andrew F; Altschul, Stephen F
2016-05-01
We describe a Bayesian Markov chain Monte Carlo (MCMC) sampler for protein multiple sequence alignment (MSA) that, as implemented in the program GISMO and applied to large numbers of diverse sequences, is more accurate than the popular MSA programs MUSCLE, MAFFT, Clustal-Ω and Kalign. Features of GISMO central to its performance are: (i) It employs a "top-down" strategy with a favorable asymptotic time complexity that first identifies regions generally shared by all the input sequences, and then realigns closely related subgroups in tandem. (ii) It infers position-specific gap penalties that favor insertions or deletions (indels) within each sequence at alignment positions in which indels are invoked in other sequences. This favors the placement of insertions between conserved blocks, which can be understood as making up the proteins' structural core. (iii) It uses a Bayesian statistical measure of alignment quality based on the minimum description length principle and on Dirichlet mixture priors. Consequently, GISMO aligns sequence regions only when statistically justified. This is unlike methods based on the ad hoc, but widely used, sum-of-the-pairs scoring system, which will align random sequences. (iv) It defines a system for exploring alignment space that provides natural avenues for further experimentation through the development of new sampling strategies for more efficiently escaping from suboptimal traps. GISMO's superior performance is illustrated using 408 protein sets containing, on average, 235 sequences. These sets correspond to NCBI Conserved Domain Database alignments, which have been manually curated in the light of available crystal structures, and thus provide a means to assess alignment accuracy. GISMO fills a different niche than other MSA programs, namely identifying and aligning a conserved domain present within a large, diverse set of full length sequences. The GISMO program is available at http://gismo.igs.umaryland.edu/. PMID:27192614
Bayesian Inference of Reticulate Phylogenies under the Multispecies Network Coalescent.
Wen, Dingqiao; Yu, Yun; Nakhleh, Luay
2016-05-01
The multispecies coalescent (MSC) is a statistical framework that models how gene genealogies grow within the branches of a species tree. The field of computational phylogenetics has witnessed an explosion in the development of methods for species tree inference under MSC, owing mainly to the accumulating evidence of incomplete lineage sorting in phylogenomic analyses. However, the evolutionary history of a set of genomes, or species, could be reticulate due to the occurrence of evolutionary processes such as hybridization or horizontal gene transfer. We report on a novel method for Bayesian inference of genome and species phylogenies under the multispecies network coalescent (MSNC). This framework models gene evolution within the branches of a phylogenetic network, thus incorporating reticulate evolutionary processes, such as hybridization, in addition to incomplete lineage sorting. As phylogenetic networks with different numbers of reticulation events correspond to points of different dimensions in the space of models, we devise a reversible-jump Markov chain Monte Carlo (RJMCMC) technique for sampling the posterior distribution of phylogenetic networks under MSNC. We implemented the methods in the publicly available, open-source software package PhyloNet and studied their performance on simulated and biological data. The work extends the reach of Bayesian inference to phylogenetic networks and enables new evolutionary analyses that account for reticulation. PMID:27144273
Bayesian Inference and Forecasting in the Stationary Bilinear Model
Roberto Leon-Gonzalez; Fuyu Yang
2014-01-01
A stationary bilinear (SB) model can be used to describe processes with a time-varying degree of persistence that depends on past shocks. An example of such a process is inflation. This study develops methods for Bayesian inference, model comparison, and forecasting in the SB model. Using monthly U.K. inflation data, we find that the SB model outperforms the random walk and first order autoregressive AR(1) models in terms of root mean squared forecast errors for both the one-step-ahead and th...
Progress on Bayesian Inference of the Fast Ion Distribution Function
Stagner, L.; Heidbrink, W.W,; Chen, X.;
2013-01-01
The fast-ion distribution function (DF) has a complicated dependence on several phase-space variables. The standard analysis procedure in energetic particle research is to compute the DF theoretically, use that DF in forward modeling to predict diagnostic signals, then compare with measured data....... However, when theory and experiment disagree (for one or more diagnostics), it is unclear how to proceed. Bayesian statistics provides a framework to infer the DF, quantify errors, and reconcile discrepant diagnostic measurements. Diagnostic errors and weight functions that describe the phase space...
Towards Bayesian Inference of the Fast-Ion Distribution Function
Stagner, L.; Heidbrink, W.W.; Salewski, Mirko
2012-01-01
The fast-ion distribution function (DF) has a complicated dependence on several phase-space variables. The standard analysis procedure in energetic particle research is to compute the DF theoretically, use that DF in forward modeling to predict diagnostic signals, then compare with measured data....... However, when theory and experiment disagree (for one or more diagnostics), it is unclear how to proceed. Bayesian statistics provides a framework to infer the DF, quantify errors, and reconcile discrepant diagnostic measurements. Diagnostic errors and ``weight functions" that describe the phase space...
Structural damage identification using piezoelectric impedance and Bayesian inference
Shuai, Q.; Zhou, K.; Tang, J.
2015-04-01
Structural damage identification is a challenging subject in the structural health monitoring research. The piezoelectric impedance-based damage identification, which usually utilizes the matrix inverse-based optimization, may in theory identify the damage location and damage severity. However, the sensitivity matrix is oftentimes ill-conditioned in practice, since the number of unknowns may far exceed the useful measurements/inputs. In this research, a new method based on intelligent inference framework for damage identification is presented. Bayesian inference is used to directly predict damage location and severity using impedance measurement through forward prediction and comparison. Gaussian process is employed to enrich the forward analysis result, thereby reducing computational cost. Case study is carried out to illustrate the identification performance.
GPU Computing in Bayesian Inference of Realized Stochastic Volatility Model
The realized stochastic volatility (RSV) model that utilizes the realized volatility as additional information has been proposed to infer volatility of financial time series. We consider the Bayesian inference of the RSV model by the Hybrid Monte Carlo (HMC) algorithm. The HMC algorithm can be parallelized and thus performed on the GPU for speedup. The GPU code is developed with CUDA Fortran. We compare the computational time in performing the HMC algorithm on GPU (GTX 760) and CPU (Intel i7-4770 3.4GHz) and find that the GPU can be up to 17 times faster than the CPU. We also code the program with OpenACC and find that appropriate coding can achieve the similar speedup with CUDA Fortran
Bayesian inference of population size history from multiple loci
Drummond Alexei J
2008-10-01
Full Text Available Abstract Background Effective population size (Ne is related to genetic variability and is a basic parameter in many models of population genetics. A number of methods for inferring current and past population sizes from genetic data have been developed since JFC Kingman introduced the n-coalescent in 1982. Here we present the Extended Bayesian Skyline Plot, a non-parametric Bayesian Markov chain Monte Carlo algorithm that extends a previous coalescent-based method in several ways, including the ability to analyze multiple loci. Results Through extensive simulations we show the accuracy and limitations of inferring population size as a function of the amount of data, including recovering information about evolutionary bottlenecks. We also analyzed two real data sets to demonstrate the behavior of the new method; a single gene Hepatitis C virus data set sampled from Egypt and a 10 locus Drosophila ananassae data set representing 16 different populations. Conclusion The results demonstrate the essential role of multiple loci in recovering population size dynamics. Multi-locus data from a small number of individuals can precisely recover past bottlenecks in population size which can not be characterized by analysis of a single locus. We also demonstrate that sequence data quality is important because even moderate levels of sequencing errors result in a considerable decrease in estimation accuracy for realistic levels of population genetic variability.
Bayesian Inference for Signal-Based Seismic Monitoring
Moore, D.
2015-12-01
Traditional seismic monitoring systems rely on discrete detections produced by station processing software, discarding significant information present in the original recorded signal. SIG-VISA (Signal-based Vertically Integrated Seismic Analysis) is a system for global seismic monitoring through Bayesian inference on seismic signals. By modeling signals directly, our forward model is able to incorporate a rich representation of the physics underlying the signal generation process, including source mechanisms, wave propagation, and station response. This allows inference in the model to recover the qualitative behavior of recent geophysical methods including waveform matching and double-differencing, all as part of a unified Bayesian monitoring system that simultaneously detects and locates events from a global network of stations. We demonstrate recent progress in scaling up SIG-VISA to efficiently process the data stream of global signals recorded by the International Monitoring System (IMS), including comparisons against existing processing methods that show increased sensitivity from our signal-based model and in particular the ability to locate events (including aftershock sequences that can tax analyst processing) precisely from waveform correlation effects. We also provide a Bayesian analysis of an alleged low-magnitude event near the DPRK test site in May 2010 [1] [2], investigating whether such an event could plausibly be detected through automated processing in a signal-based monitoring system. [1] Zhang, Miao and Wen, Lianxing. "Seismological Evidence for a Low-Yield Nuclear Test on 12 May 2010 in North Korea". Seismological Research Letters, January/February 2015. [2] Richards, Paul. "A Seismic Event in North Korea on 12 May 2010". CTBTO SnT 2015 oral presentation, video at https://video-archive.ctbto.org/index.php/kmc/preview/partner_id/103/uiconf_id/4421629/entry_id/0_ymmtpps0/delivery/http
Bayesian inference for identifying interaction rules in moving animal groups.
Mann, Richard P
2011-01-01
The emergence of similar collective patterns from different self-propelled particle models of animal groups points to a restricted set of "universal" classes for these patterns. While universality is interesting, it is often the fine details of animal interactions that are of biological importance. Universality thus presents a challenge to inferring such interactions from macroscopic group dynamics since these can be consistent with many underlying interaction models. We present a Bayesian framework for learning animal interaction rules from fine scale recordings of animal movements in swarms. We apply these techniques to the inverse problem of inferring interaction rules from simulation models, showing that parameters can often be inferred from a small number of observations. Our methodology allows us to quantify our confidence in parameter fitting. For example, we show that attraction and alignment terms can be reliably estimated when animals are milling in a torus shape, while interaction radius cannot be reliably measured in such a situation. We assess the importance of rate of data collection and show how to test different models, such as topological and metric neighbourhood models. Taken together our results both inform the design of experiments on animal interactions and suggest how these data should be best analysed. PMID:21829657
Bayesian inference for identifying interaction rules in moving animal groups.
Richard P Mann
Full Text Available The emergence of similar collective patterns from different self-propelled particle models of animal groups points to a restricted set of "universal" classes for these patterns. While universality is interesting, it is often the fine details of animal interactions that are of biological importance. Universality thus presents a challenge to inferring such interactions from macroscopic group dynamics since these can be consistent with many underlying interaction models. We present a Bayesian framework for learning animal interaction rules from fine scale recordings of animal movements in swarms. We apply these techniques to the inverse problem of inferring interaction rules from simulation models, showing that parameters can often be inferred from a small number of observations. Our methodology allows us to quantify our confidence in parameter fitting. For example, we show that attraction and alignment terms can be reliably estimated when animals are milling in a torus shape, while interaction radius cannot be reliably measured in such a situation. We assess the importance of rate of data collection and show how to test different models, such as topological and metric neighbourhood models. Taken together our results both inform the design of experiments on animal interactions and suggest how these data should be best analysed.
Inference of Gene Regulatory Network Based on Local Bayesian Networks
Liu, Fei; Zhang, Shao-Wu; Guo, Wei-Feng; Chen, Luonan
2016-01-01
The inference of gene regulatory networks (GRNs) from expression data can mine the direct regulations among genes and gain deep insights into biological processes at a network level. During past decades, numerous computational approaches have been introduced for inferring the GRNs. However, many of them still suffer from various problems, e.g., Bayesian network (BN) methods cannot handle large-scale networks due to their high computational complexity, while information theory-based methods cannot identify the directions of regulatory interactions and also suffer from false positive/negative problems. To overcome the limitations, in this work we present a novel algorithm, namely local Bayesian network (LBN), to infer GRNs from gene expression data by using the network decomposition strategy and false-positive edge elimination scheme. Specifically, LBN algorithm first uses conditional mutual information (CMI) to construct an initial network or GRN, which is decomposed into a number of local networks or GRNs. Then, BN method is employed to generate a series of local BNs by selecting the k-nearest neighbors of each gene as its candidate regulatory genes, which significantly reduces the exponential search space from all possible GRN structures. Integrating these local BNs forms a tentative network or GRN by performing CMI, which reduces redundant regulations in the GRN and thus alleviates the false positive problem. The final network or GRN can be obtained by iteratively performing CMI and local BN on the tentative network. In the iterative process, the false or redundant regulations are gradually removed. When tested on the benchmark GRN datasets from DREAM challenge as well as the SOS DNA repair network in E.coli, our results suggest that LBN outperforms other state-of-the-art methods (ARACNE, GENIE3 and NARROMI) significantly, with more accurate and robust performance. In particular, the decomposition strategy with local Bayesian networks not only effectively reduce
Inference of Gene Regulatory Network Based on Local Bayesian Networks.
Liu, Fei; Zhang, Shao-Wu; Guo, Wei-Feng; Wei, Ze-Gang; Chen, Luonan
2016-08-01
The inference of gene regulatory networks (GRNs) from expression data can mine the direct regulations among genes and gain deep insights into biological processes at a network level. During past decades, numerous computational approaches have been introduced for inferring the GRNs. However, many of them still suffer from various problems, e.g., Bayesian network (BN) methods cannot handle large-scale networks due to their high computational complexity, while information theory-based methods cannot identify the directions of regulatory interactions and also suffer from false positive/negative problems. To overcome the limitations, in this work we present a novel algorithm, namely local Bayesian network (LBN), to infer GRNs from gene expression data by using the network decomposition strategy and false-positive edge elimination scheme. Specifically, LBN algorithm first uses conditional mutual information (CMI) to construct an initial network or GRN, which is decomposed into a number of local networks or GRNs. Then, BN method is employed to generate a series of local BNs by selecting the k-nearest neighbors of each gene as its candidate regulatory genes, which significantly reduces the exponential search space from all possible GRN structures. Integrating these local BNs forms a tentative network or GRN by performing CMI, which reduces redundant regulations in the GRN and thus alleviates the false positive problem. The final network or GRN can be obtained by iteratively performing CMI and local BN on the tentative network. In the iterative process, the false or redundant regulations are gradually removed. When tested on the benchmark GRN datasets from DREAM challenge as well as the SOS DNA repair network in E.coli, our results suggest that LBN outperforms other state-of-the-art methods (ARACNE, GENIE3 and NARROMI) significantly, with more accurate and robust performance. In particular, the decomposition strategy with local Bayesian networks not only effectively reduce
Bayesian inference on the sphere beyond statistical isotropy
Das, Santanu; Souradeep, Tarun
2015-01-01
We present a general method for Bayesian inference of the underlying covariance structure of random fields on a sphere. We employ the Bipolar Spherical Harmonic (BipoSH) representation of general covariance structure on the sphere. We illustrate the efficacy of the method as a principled approach to assess violation of statistical isotropy (SI) in the sky maps of Cosmic Microwave Background (CMB) fluctuations. SI violation in observed CMB maps arise due to known physical effects such as Doppler boost and weak lensing; yet unknown theoretical possibilities like cosmic topology and subtle violations of the cosmological principle, as well as, expected observational artefacts of scanning the sky with a non-circular beam, masking, foreground residuals, anisotropic noise, etc. We explicitly demonstrate the recovery of the input SI violation signals with their full statistics in simulated CMB maps. Our formalism easily adapts to exploring parametric physical models with non-SI covariance, as we illustrate for the in...
Bayesian Inference of Natural Rankings in Incomplete Competition Networks
Park, Juyong
2013-01-01
Competition between a complex system's constituents and a corresponding reward mechanism based on it have profound influence on the functioning, stability, and evolution of the system. But determining the dominance hierarchy or ranking among the constituent parts from the strongest to the weakest -- essential in determining reward or penalty -- is almost always an ambiguous task due to the incomplete nature of competition networks. Here we introduce ``Natural Ranking," a desirably unambiguous ranking method applicable to a complete (full) competition network, and formulate an analytical model based on the Bayesian formula inferring the expected mean and error of the natural ranking of nodes from an incomplete network. We investigate its potential and uses in solving issues in ranking by applying to a real-world competition network of economic and social importance.
Bayesian large-scale structure inference and cosmic web analysis
Leclercq, Florent
2015-01-01
Surveys of the cosmic large-scale structure carry opportunities for building and testing cosmological theories about the origin and evolution of the Universe. This endeavor requires appropriate data assimilation tools, for establishing the contact between survey catalogs and models of structure formation. In this thesis, we present an innovative statistical approach for the ab initio simultaneous analysis of the formation history and morphology of the cosmic web: the BORG algorithm infers the primordial density fluctuations and produces physical reconstructions of the dark matter distribution that underlies observed galaxies, by assimilating the survey data into a cosmological structure formation model. The method, based on Bayesian probability theory, provides accurate means of uncertainty quantification. We demonstrate the application of BORG to the Sloan Digital Sky Survey data and describe the primordial and late-time large-scale structure in the observed volume. We show how the approach has led to the fi...
Unsupervised Transient Light Curve Analysis Via Hierarchical Bayesian Inference
Sanders, Nathan; Soderberg, Alicia
2014-01-01
Historically, light curve studies of supernovae (SNe) and other transient classes have focused on individual objects with copious and high signal-to-noise observations. In the nascent era of wide field transient searches, objects with detailed observations are decreasing as a fraction of the overall known SN population, and this strategy sacrifices the majority of the information contained in the data about the underlying population of transients. A population level modeling approach, simultaneously fitting all available observations of objects in a transient sub-class of interest, fully mines the data to infer the properties of the population and avoids certain systematic biases. We present a novel hierarchical Bayesian statistical model for population level modeling of transient light curves, and discuss its implementation using an efficient Hamiltonian Monte Carlo technique. As a test case, we apply this model to the Type IIP SN sample from the Pan-STARRS1 Medium Deep Survey, consisting of 18,837 photometr...
Metainference: A Bayesian Inference Method for Heterogeneous Systems
Bonomi, Massimiliano; Cavalli, Andrea; Vendruscolo, Michele
2015-01-01
Modelling a complex system is almost invariably a challenging task. The incorporation of experimental observations can be used to improve the quality of a model, and thus to obtain better predictions about the behavior of the corresponding system. This approach, however, is affected by a variety of different errors, especially when a system populates simultaneously an ensemble of different states and experimental data are measured as averages over such states. To address this problem we present a method, called metainference, that combines Bayesian inference, which is a powerful strategy to deal with errors in experimental measurements, with the maximum entropy principle, which represents a rigorous approach to deal with experimental measurements averaged over multiple states. To illustrate the method we present its application to the determination of an ensemble of structures corresponding to the thermal fluctuations of a protein molecule. Metainference thus provides an approach to model complex systems with...
Bayesian inference of mass segregation of open clusters
Shao, Zhengyi; Chen, Li; Lin, Chien-Cheng; Zhong, Jing; Hou, Jinliang
2015-08-01
Based on the Bayesian inference (BI) method, the mixture-modeling approach is improved to combine all kinematic data, including the coordinative position, proper motion (PM) and radial velocity (RV), to separate the motion of the cluster from field stars in its area, as well as to describe the intrinsic kinematic status. Meanwhile, the membership probabilities of individual stars are determined as by product results. This method has been testified by simulation of toy models and it was found that the joint usage of multiple kinematic data can significantly reduce the missing rate of membership determination, say from ~15% for single data type to 1% for using all position, proper motion and radial velocity data.By combining kinematic data from multiple sources of photometric and redshift surveys, such as WIYN and APOGEE, M67 and NGC188 are revisited. Mass segregation is identified clearly for both of these two old open clusters, either in position or in PM spaces, since the Bayesian evidence (BE) of the model, which includes the segregation parameters, is much larger than that without it. The ongoing work is applying this method to the LAMOST released data which contains a large amount of RVs cover ~200 nearby open clusters. If the coming GAIA data can be used, the accuracy of tangential velocity will be largely improved and the intrinsic kinematics of open clusters can be well investigated, though they are usually less than 1 km/s.
Bayesian Fusion Algorithm for Inferring Trust in Wireless Sensor Networks
Mohammad Momani
2010-07-01
Full Text Available This paper introduces a new Bayesian fusion algorithm to combine more than one trust component (data trust and communication trust to infer the overall trust between nodes. This research work proposes that one trust component is not enough when deciding on whether or not to trust a specific node in a wireless sensor network. This paper discusses and analyses the results from the communication trust component (binary and the data trust component (continuous and proves that either component by itself, can mislead the network and eventually cause a total breakdown of the network. As a result of this, new algorithms are needed to combine more than one trust component to infer the overall trust. The proposed algorithm is simple and generic as it allows trust components to be added and deleted easily. Simulation results demonstrate that a node is highly trustworthy provided that both trust components simultaneously confirm its trustworthiness and conversely, a node is highly untrustworthy if its untrustworthiness is asserted by both components.
vs. a polynomial chaos-based MCMC
Siripatana, Adil
2014-08-01
Bayesian Inference of Manning\\'s n coefficient in a Storm Surge Model Framework: comparison between Kalman lter and polynomial based method Adil Siripatana Conventional coastal ocean models solve the shallow water equations, which describe the conservation of mass and momentum when the horizontal length scale is much greater than the vertical length scale. In this case vertical pressure gradients in the momentum equations are nearly hydrostatic. The outputs of coastal ocean models are thus sensitive to the bottom stress terms de ned through the formulation of Manning\\'s n coefficients. This thesis considers the Bayesian inference problem of the Manning\\'s n coefficient in the context of storm surge based on the coastal ocean ADCIRC model. In the first part of the thesis, we apply an ensemble-based Kalman filter, the singular evolutive interpolated Kalman (SEIK) filter to estimate both a constant Manning\\'s n coefficient and a 2-D parameterized Manning\\'s coefficient on one ideal and one of more realistic domain using observation system simulation experiments (OSSEs). We study the sensitivity of the system to the ensemble size. we also access the benefits from using an in ation factor on the filter performance. To study the limitation of the Guassian restricted assumption on the SEIK lter, 5 we also implemented in the second part of this thesis a Markov Chain Monte Carlo (MCMC) method based on a Generalized Polynomial chaos (gPc) approach for the estimation of the 1-D and 2-D Mannning\\'s n coe cient. The gPc is used to build a surrogate model that imitate the ADCIRC model in order to make the computational cost of implementing the MCMC with the ADCIRC model reasonable. We evaluate the performance of the MCMC-gPc approach and study its robustness to di erent OSSEs scenario. we also compare its estimates with those resulting from SEIK in term of parameter estimates and full distributions. we present a full analysis of the solution of these two methods, of the
Rajabi, Mohammad Mahdi; Ataie-Ashtiani, Behzad
2016-05-01
Bayesian inference has traditionally been conceived as the proper framework for the formal incorporation of expert knowledge in parameter estimation of groundwater models. However, conventional Bayesian inference is incapable of taking into account the imprecision essentially embedded in expert provided information. In order to solve this problem, a number of extensions to conventional Bayesian inference have been introduced in recent years. One of these extensions is 'fuzzy Bayesian inference' which is the result of integrating fuzzy techniques into Bayesian statistics. Fuzzy Bayesian inference has a number of desirable features which makes it an attractive approach for incorporating expert knowledge in the parameter estimation process of groundwater models: (1) it is well adapted to the nature of expert provided information, (2) it allows to distinguishably model both uncertainty and imprecision, and (3) it presents a framework for fusing expert provided information regarding the various inputs of the Bayesian inference algorithm. However an important obstacle in employing fuzzy Bayesian inference in groundwater numerical modeling applications is the computational burden, as the required number of numerical model simulations often becomes extremely exhaustive and often computationally infeasible. In this paper, a novel approach of accelerating the fuzzy Bayesian inference algorithm is proposed which is based on using approximate posterior distributions derived from surrogate modeling, as a screening tool in the computations. The proposed approach is first applied to a synthetic test case of seawater intrusion (SWI) in a coastal aquifer. It is shown that for this synthetic test case, the proposed approach decreases the number of required numerical simulations by an order of magnitude. Then the proposed approach is applied to a real-world test case involving three-dimensional numerical modeling of SWI in Kish Island, located in the Persian Gulf. An expert
Bayesian inference for partially identified models exploring the limits of limited data
Gustafson, Paul
2015-01-01
Introduction Identification What Is against Us? What Is for Us? Some Simple Examples of Partially Identified ModelsThe Road Ahead The Structure of Inference in Partially Identified Models Bayesian Inference The Structure of Posterior Distributions in PIMs Computational Strategies Strength of Bayesian Updating, Revisited Posterior MomentsCredible Intervals Evaluating the Worth of Inference Partial Identification versus Model Misspecification The Siren Call of Identification Comp
On Russian Roulette Estimates for Bayesian Inference with Doubly-Intractable Likelihoods
Lyne, Anne-Marie; Girolami, Mark; Atchadé, Yves; Strathmann, Heiko; Simpson, Daniel
2013-01-01
A large number of statistical models are “doubly-intractable”: the likelihood normalising term, which is a function of the model parameters, is intractable, as well as the marginal likelihood (model evidence). This means that standard inference techniques to sample from the posterior, such as Markov chain Monte Carlo (MCMC), cannot be used. Examples include, but are not confined to, massive Gaussian Markov random fields, autologistic models and Exponential random graph models. A number of app...
Bayesian Inference for Functional Dynamics Exploring in fMRI Data
Xuan Guo
2016-01-01
Full Text Available This paper aims to review state-of-the-art Bayesian-inference-based methods applied to functional magnetic resonance imaging (fMRI data. Particularly, we focus on one specific long-standing challenge in the computational modeling of fMRI datasets: how to effectively explore typical functional interactions from fMRI time series and the corresponding boundaries of temporal segments. Bayesian inference is a method of statistical inference which has been shown to be a powerful tool to encode dependence relationships among the variables with uncertainty. Here we provide an introduction to a group of Bayesian-inference-based methods for fMRI data analysis, which were designed to detect magnitude or functional connectivity change points and to infer their functional interaction patterns based on corresponding temporal boundaries. We also provide a comparison of three popular Bayesian models, that is, Bayesian Magnitude Change Point Model (BMCPM, Bayesian Connectivity Change Point Model (BCCPM, and Dynamic Bayesian Variable Partition Model (DBVPM, and give a summary of their applications. We envision that more delicate Bayesian inference models will be emerging and play increasingly important roles in modeling brain functions in the years to come.
Hernández, Mario R.; Francés, Félix
2015-04-01
One phase of the hydrological models implementation process, significantly contributing to the hydrological predictions uncertainty, is the calibration phase in which values of the unknown model parameters are tuned by optimizing an objective function. An unsuitable error model (e.g. Standard Least Squares or SLS) introduces noise into the estimation of the parameters. The main sources of this noise are the input errors and the hydrological model structural deficiencies. Thus, the biased calibrated parameters cause the divergence model phenomenon, where the errors variance of the (spatially and temporally) forecasted flows far exceeds the errors variance in the fitting period, and provoke the loss of part or all of the physical meaning of the modeled processes. In other words, yielding a calibrated hydrological model which works well, but not for the right reasons. Besides, an unsuitable error model yields a non-reliable predictive uncertainty assessment. Hence, with the aim of prevent all these undesirable effects, this research focuses on the Bayesian joint inference (BJI) of both the hydrological and error model parameters, considering a general additive (GA) error model that allows for correlation, non-stationarity (in variance and bias) and non-normality of model residuals. As hydrological model, it has been used a conceptual distributed model called TETIS, with a particular split structure of the effective model parameters. Bayesian inference has been performed with the aid of a Markov Chain Monte Carlo (MCMC) algorithm called Dream-ZS. MCMC algorithm quantifies the uncertainty of the hydrological and error model parameters by getting the joint posterior probability distribution, conditioned on the observed flows. The BJI methodology is a very powerful and reliable tool, but it must be used correctly this is, if non-stationarity in errors variance and bias is modeled, the Total Laws must be taken into account. The results of this research show that the
Bayesian analysis of Markov point processes
Berthelsen, Kasper Klitgaard; Møller, Jesper
2006-01-01
Recently Møller, Pettitt, Berthelsen and Reeves introduced a new MCMC methodology for drawing samples from a posterior distribution when the likelihood function is only specified up to a normalising constant. We illustrate the method in the setting of Bayesian inference for Markov point processes...
Bayesian Inference for Neighborhood Filters With Application in Denoising.
Huang, Chao-Tsung
2015-11-01
Range-weighted neighborhood filters are useful and popular for their edge-preserving property and simplicity, but they are originally proposed as intuitive tools. Previous works needed to connect them to other tools or models for indirect property reasoning or parameter estimation. In this paper, we introduce a unified empirical Bayesian framework to do both directly. A neighborhood noise model is proposed to reason and infer the Yaroslavsky, bilateral, and modified non-local means filters by joint maximum a posteriori and maximum likelihood estimation. Then, the essential parameter, range variance, can be estimated via model fitting to the empirical distribution of an observable chi scale mixture variable. An algorithm based on expectation-maximization and quasi-Newton optimization is devised to perform the model fitting efficiently. Finally, we apply this framework to the problem of color-image denoising. A recursive fitting and filtering scheme is proposed to improve the image quality. Extensive experiments are performed for a variety of configurations, including different kernel functions, filter types and support sizes, color channel numbers, and noise types. The results show that the proposed framework can fit noisy images well and the range variance can be estimated successfully and efficiently. PMID:26259244
Bayesian inference and life testing plans for generalized exponential distribution
KUNDU; Debasis; PRADHAN; Biswabrata
2009-01-01
Recently generalized exponential distribution has received considerable attentions.In this paper,we deal with the Bayesian inference of the unknown parameters of the progressively censored generalized exponential distribution.It is assumed that the scale and the shape parameters have independent gamma priors.The Bayes estimates of the unknown parameters cannot be obtained in the closed form.Lindley’s approximation and importance sampling technique have been suggested to compute the approximate Bayes estimates.Markov Chain Monte Carlo method has been used to compute the approximate Bayes estimates and also to construct the highest posterior density credible intervals.We also provide different criteria to compare two different sampling schemes and hence to ?nd the optimal sampling schemes.It is observed that ?nding the optimum censoring procedure is a computationally expensive process.And we have recommended to use the sub-optimal censoring procedure,which can be obtained very easily.Monte Carlo simulations are performed to compare the performances of the different methods and one data analysis has been performed for illustrative purposes.
Metainference: A Bayesian inference method for heterogeneous systems.
Bonomi, Massimiliano; Camilloni, Carlo; Cavalli, Andrea; Vendruscolo, Michele
2016-01-01
Modeling a complex system is almost invariably a challenging task. The incorporation of experimental observations can be used to improve the quality of a model and thus to obtain better predictions about the behavior of the corresponding system. This approach, however, is affected by a variety of different errors, especially when a system simultaneously populates an ensemble of different states and experimental data are measured as averages over such states. To address this problem, we present a Bayesian inference method, called "metainference," that is able to deal with errors in experimental measurements and with experimental measurements averaged over multiple states. To achieve this goal, metainference models a finite sample of the distribution of models using a replica approach, in the spirit of the replica-averaging modeling based on the maximum entropy principle. To illustrate the method, we present its application to a heterogeneous model system and to the determination of an ensemble of structures corresponding to the thermal fluctuations of a protein molecule. Metainference thus provides an approach to modeling complex systems with heterogeneous components and interconverting between different states by taking into account all possible sources of errors. PMID:26844300
Bayesian Inference for NASA Probabilistic Risk and Reliability Analysis
Dezfuli, Homayoon; Kelly, Dana; Smith, Curtis; Vedros, Kurt; Galyean, William
2009-01-01
This document, Bayesian Inference for NASA Probabilistic Risk and Reliability Analysis, is intended to provide guidelines for the collection and evaluation of risk and reliability-related data. It is aimed at scientists and engineers familiar with risk and reliability methods and provides a hands-on approach to the investigation and application of a variety of risk and reliability data assessment methods, tools, and techniques. This document provides both: A broad perspective on data analysis collection and evaluation issues. A narrow focus on the methods to implement a comprehensive information repository. The topics addressed herein cover the fundamentals of how data and information are to be used in risk and reliability analysis models and their potential role in decision making. Understanding these topics is essential to attaining a risk informed decision making environment that is being sought by NASA requirements and procedures such as 8000.4 (Agency Risk Management Procedural Requirements), NPR 8705.05 (Probabilistic Risk Assessment Procedures for NASA Programs and Projects), and the System Safety requirements of NPR 8715.3 (NASA General Safety Program Requirements).
Fast, fully Bayesian spatiotemporal inference for fMRI data.
Musgrove, Donald R; Hughes, John; Eberly, Lynn E
2016-04-01
We propose a spatial Bayesian variable selection method for detecting blood oxygenation level dependent activation in functional magnetic resonance imaging (fMRI) data. Typical fMRI experiments generate large datasets that exhibit complex spatial and temporal dependence. Fitting a full statistical model to such data can be so computationally burdensome that many practitioners resort to fitting oversimplified models, which can lead to lower quality inference. We develop a full statistical model that permits efficient computation. Our approach eases the computational burden in two ways. We partition the brain into 3D parcels, and fit our model to the parcels in parallel. Voxel-level activation within each parcel is modeled as regressions located on a lattice. Regressors represent the magnitude of change in blood oxygenation in response to a stimulus, while a latent indicator for each regressor represents whether the change is zero or non-zero. A sparse spatial generalized linear mixed model captures the spatial dependence among indicator variables within a parcel and for a given stimulus. The sparse SGLMM permits considerably more efficient computation than does the spatial model typically employed in fMRI. Through simulation we show that our parcellation scheme performs well in various realistic scenarios. Importantly, indicator variables on the boundary between parcels do not exhibit edge effects. We conclude by applying our methodology to data from a task-based fMRI experiment. PMID:26553916
Markov Model of Wind Power Time Series UsingBayesian Inference of Transition Matrix
Chen, Peiyuan; Berthelsen, Kasper Klitgaard; Bak-Jensen, Birgitte; Chen, Zhe
2009-01-01
This paper proposes to use Bayesian inference of transition matrix when developing a discrete Markov model of a wind speed/power time series and 95% credible interval for the model verification. The Dirichlet distribution is used as a conjugate prior for the transition matrix. Three discrete Markov models are compared, i.e. the basic Markov model, the Bayesian Markov model and the birth-and-death Markov model. The proposed Bayesian Markov model shows the best accuracy in modeling the autocorr...
Mocapy++ - A toolkit for inference and learning in dynamic Bayesian networks
Paluszewski, Martin; Hamelryck, Thomas Wim
2010-01-01
Background Mocapy++ is a toolkit for parameter learning and inference in dynamic Bayesian networks (DBNs). It supports a wide range of DBN architectures and probability distributions, including distributions from directional statistics (the statistics of angles, directions and orientations...
VIGoR: Variational Bayesian Inference for Genome-Wide Regression
Onogi, Akio; Iwata, Hiroyoshi
2016-01-01
Genome-wide regression using a number of genome-wide markers as predictors is now widely used for genome-wide association mapping and genomic prediction. We developed novel software for genome-wide regression which we named VIGoR (variational Bayesian inference for genome-wide regression). Variational Bayesian inference is computationally much faster than widely used Markov chain Monte Carlo algorithms. VIGoR implements seven regression methods, and is provided as a command line program packa...
UNSUPERVISED TRANSIENT LIGHT CURVE ANALYSIS VIA HIERARCHICAL BAYESIAN INFERENCE
Sanders, N. E.; Soderberg, A. M. [Harvard-Smithsonian Center for Astrophysics, 60 Garden Street, Cambridge, MA 02138 (United States); Betancourt, M., E-mail: nsanders@cfa.harvard.edu [Department of Statistics, University of Warwick, Coventry CV4 7AL (United Kingdom)
2015-02-10
Historically, light curve studies of supernovae (SNe) and other transient classes have focused on individual objects with copious and high signal-to-noise observations. In the nascent era of wide field transient searches, objects with detailed observations are decreasing as a fraction of the overall known SN population, and this strategy sacrifices the majority of the information contained in the data about the underlying population of transients. A population level modeling approach, simultaneously fitting all available observations of objects in a transient sub-class of interest, fully mines the data to infer the properties of the population and avoids certain systematic biases. We present a novel hierarchical Bayesian statistical model for population level modeling of transient light curves, and discuss its implementation using an efficient Hamiltonian Monte Carlo technique. As a test case, we apply this model to the Type IIP SN sample from the Pan-STARRS1 Medium Deep Survey, consisting of 18,837 photometric observations of 76 SNe, corresponding to a joint posterior distribution with 9176 parameters under our model. Our hierarchical model fits provide improved constraints on light curve parameters relevant to the physical properties of their progenitor stars relative to modeling individual light curves alone. Moreover, we directly evaluate the probability for occurrence rates of unseen light curve characteristics from the model hyperparameters, addressing observational biases in survey methodology. We view this modeling framework as an unsupervised machine learning technique with the ability to maximize scientific returns from data to be collected by future wide field transient searches like LSST.
Bayesian inference of the demographic history of chimpanzees.
Wegmann, Daniel; Excoffier, Laurent
2010-06-01
Due to an almost complete absence of fossil record, the evolutionary history of chimpanzees has only been studied recently on the basis of genetic data. Although the general topology of the chimpanzee phylogeny is well established, uncertainties remain concerning the size of current and past populations, the occurrence of bottlenecks or population expansions, or about divergence times and migrations rates between subspecies. Here, we present a novel attempt at globally inferring the detailed evolution of the Pan genus based on approximate Bayesian computation, an approach preferentially applied to complex models where the likelihood cannot be computed analytically. Based on two microsatellite and DNA sequence data sets and adjusting simulated data for local levels of inbreeding and patterns of missing data, we find support for several new features of chimpanzee evolution as compared with previous studies based on smaller data sets and simpler evolutionary models. We find that the central chimpanzees are certainly the oldest population of all P. troglodytes subspecies and that the other two P. t. subspecies diverged from the central chimpanzees by founder events. We also find an older divergence time (1.6 million years [My]) between common chimpanzee and Bonobos than previous studies (0.9-1.3 My), but this divergence appears to have been very progressive with the maintenance of relatively high levels of gene flow between the ancestral chimpanzee population and the Bonobos. Finally, we could also confirm the existence of strong unidirectional gene flow from the western into the central chimpanzee. These results show that interesting and innovative features of chimpanzee history emerge when considering their whole evolutionary history in a single analysis, rather than relying on simpler models involving several comparisons of pairs of populations. PMID:20118191
Bayesian inference of nonlinear unsteady aerodynamics from aeroelastic limit cycle oscillations
Sandhu, Rimple; Poirel, Dominique; Pettit, Chris; Khalil, Mohammad; Sarkar, Abhijit
2016-07-01
A Bayesian model selection and parameter estimation algorithm is applied to investigate the influence of nonlinear and unsteady aerodynamic loads on the limit cycle oscillation (LCO) of a pitching airfoil in the transitional Reynolds number regime. At small angles of attack, laminar boundary layer trailing edge separation causes negative aerodynamic damping leading to the LCO. The fluid-structure interaction of the rigid, but elastically mounted, airfoil and nonlinear unsteady aerodynamics is represented by two coupled nonlinear stochastic ordinary differential equations containing uncertain parameters and model approximation errors. Several plausible aerodynamic models with increasing complexity are proposed to describe the aeroelastic system leading to LCO. The likelihood in the posterior parameter probability density function (pdf) is available semi-analytically using the extended Kalman filter for the state estimation of the coupled nonlinear structural and unsteady aerodynamic model. The posterior parameter pdf is sampled using a parallel and adaptive Markov Chain Monte Carlo (MCMC) algorithm. The posterior probability of each model is estimated using the Chib-Jeliazkov method that directly uses the posterior MCMC samples for evidence (marginal likelihood) computation. The Bayesian algorithm is validated through a numerical study and then applied to model the nonlinear unsteady aerodynamic loads using wind-tunnel test data at various Reynolds numbers.
Elsheikh, Ahmed H.
2014-02-01
An efficient Bayesian calibration method based on the nested sampling (NS) algorithm and non-intrusive polynomial chaos method is presented. Nested sampling is a Bayesian sampling algorithm that builds a discrete representation of the posterior distributions by iteratively re-focusing a set of samples to high likelihood regions. NS allows representing the posterior probability density function (PDF) with a smaller number of samples and reduces the curse of dimensionality effects. The main difficulty of the NS algorithm is in the constrained sampling step which is commonly performed using a random walk Markov Chain Monte-Carlo (MCMC) algorithm. In this work, we perform a two-stage sampling using a polynomial chaos response surface to filter out rejected samples in the Markov Chain Monte-Carlo method. The combined use of nested sampling and the two-stage MCMC based on approximate response surfaces provides significant computational gains in terms of the number of simulation runs. The proposed algorithm is applied for calibration and model selection of subsurface flow models. © 2013.
Bayesian networks inference algorithm to implement Dempster Shafer theory in reliability analysis
This paper deals with the use of Bayesian networks to compute system reliability. The reliability analysis problem is described and the usual methods for quantitative reliability analysis are presented within a case study. Some drawbacks that justify the use of Bayesian networks are identified. The basic concepts of the Bayesian networks application to reliability analysis are introduced and a model to compute the reliability for the case study is presented. Dempster Shafer theory to treat epistemic uncertainty in reliability analysis is then discussed and its basic concepts that can be applied thanks to the Bayesian network inference algorithm are introduced. Finally, it is shown, with a numerical example, how Bayesian networks' inference algorithms compute complex system reliability and what the Dempster Shafer theory can provide to reliability analysis
Online query answering with differential privacy: a utility-driven approach using Bayesian inference
Xiao, Yonghui
2012-01-01
Data privacy issues frequently and increasingly arise for data sharing and data analysis tasks. In this paper, we study the problem of online query answering under the rigorous differential privacy model. The existing interactive mechanisms for differential privacy can only support a limited number of queries before the accumulated cost of privacy reaches a certain bound. This limitation has greatly hindered their applicability, especially in the scenario where multiple users legitimately need to pose a large number of queries. To minimize the privacy cost and extend the life span of a system, we propose a utility-driven mechanism for online query answering using Bayesian statistical inference. The key idea is to keep track of the query history and use Bayesian inference to answer a new query using previous query answers. The Bayesian inference algorithm provides both optimal point estimation and optimal interval estimation. We formally quantify the error of the inference result to determine if it satisfies t...
A Non-Parametric Bayesian Method for Inferring Hidden Causes
Wood, Frank; Griffiths, Thomas; Ghahramani, Zoubin
2012-01-01
We present a non-parametric Bayesian approach to structure learning with hidden causes. Previous Bayesian treatments of this problem define a prior over the number of hidden causes and use algorithms such as reversible jump Markov chain Monte Carlo to move between solutions. In contrast, we assume that the number of hidden causes is unbounded, but only a finite number influence observable variables. This makes it possible to use a Gibbs sampler to approximate the distribution over causal stru...
Protein NMR Structure Refinement based on Bayesian Inference
Ikeya, Teppei; Ikeda, Shiro; Kigawa, Takanori; Ito, Yutaka; Güntert, Peter
2016-03-01
Nuclear Magnetic Resonance (NMR) spectroscopy is a tool to investigate threedimensional (3D) structures and dynamics of biomacromolecules at atomic resolution in solution or more natural environments such as living cells. Since NMR data are principally only spectra with peak signals, it is required to properly deduce structural information from the sparse experimental data with their imperfections and uncertainty, and to visualize 3D conformations by NMR structure calculation. In order to efficiently analyse the data, Rieping et al. proposed a new structure calculation method based on Bayes’ theorem. We implemented a similar approach into the program CYANA with some modifications. It allows us to handle automatic NOE cross peak assignments in unambiguous and ambiguous usages, and to create a prior distribution based on a physical force field with the generalized Born implicit water model. The sampling scheme for obtaining the posterior is performed by a hybrid Monte Carlo algorithm combined with Markov chain Monte Carlo (MCMC) by the Gibbs sampler, and molecular dynamics simulation (MD) for obtaining a canonical ensemble of conformations. Since it is not trivial to search the entire function space particularly for exploring the conformational prior due to the extraordinarily large conformation space of proteins, the replica exchange method is performed, in which several MCMC calculations with different temperatures run in parallel as replicas. It is shown with simulated data or randomly deleted experimental peaks that the new structure calculation method can provide accurate structures even with less peaks, especially compared with the conventional method. In particular, it dramatically improves in-cell structures of the proteins GB1 and TTHA1718 using exclusively information obtained in living Escherichia coli (E. coli) cells.
Sythesis of MCMC and Belief Propagation
Ahn, Sungsoo [Korea Advanced Institute of Science and Technology, Daejeon (South Korea); Chertkov, Michael [Los Alamos National Lab. (LANL), Los Alamos, NM (United States); Shin, Jinwoo [Korea Advanced Institute of Science and Technology, Daejeon (South Korea)
2016-05-27
Markov Chain Monte Carlo (MCMC) and Belief Propagation (BP) are the most popular algorithms for computational inference in Graphical Models (GM). In principle, MCMC is an exact probabilistic method which, however, often suffers from exponentially slow mixing. In contrast, BP is a deterministic method, which is typically fast, empirically very successful, however in general lacking control of accuracy over loopy graphs. In this paper, we introduce MCMC algorithms correcting the approximation error of BP, i.e., we provide a way to compensate for BP errors via a consecutive BP-aware MCMC. Our framework is based on the Loop Calculus (LC) approach which allows to express the BP error as a sum of weighted generalized loops. Although the full series is computationally intractable, it is known that a truncated series, summing up all 2-regular loops, is computable in polynomial-time for planar pair-wise binary GMs and it also provides a highly accurate approximation empirically. Motivated by this, we first propose a polynomial-time approximation MCMC scheme for the truncated series of general (non-planar) pair-wise binary models. Our main idea here is to use the Worm algorithm, known to provide fast mixing in other (related) problems, and then design an appropriate rejection scheme to sample 2-regular loops. Furthermore, we also design an efficient rejection-free MCMC scheme for approximating the full series. The main novelty underlying our design is in utilizing the concept of cycle basis, which provides an efficient decomposition of the generalized loops. In essence, the proposed MCMC schemes run on transformed GM built upon the non-trivial BP solution, and our experiments show that this synthesis of BP and MCMC outperforms both direct MCMC and bare BP schemes.
The Lumiere Project: Bayesian User Modeling for Inferring the Goals and Needs of Software Users
Horvitz, Eric J.; Breese, John S.; Heckerman, David; Hovel, David; Rommelse, Koos
2013-01-01
The Lumiere Project centers on harnessing probability and utility to provide assistance to computer software users. We review work on Bayesian user models that can be employed to infer a users needs by considering a user's background, actions, and queries. Several problems were tackled in Lumiere research, including (1) the construction of Bayesian models for reasoning about the time-varying goals of computer users from their observed actions and queries, (2) gaining access to a stream of eve...
Bayesian inference of BWR model parameters by Markov chain Monte Carlo
In this paper, the Markov chain Monte Carlo approach to Bayesian inference is applied for estimating the parameters of a reduced-order model of the dynamics of a boiling water reactor system. A Bayesian updating strategy is devised to progressively refine the estimates, as newly measured data become available. Finally, the technique is used for detecting parameter changes during the system lifetime, e.g. due to component degradation
Jannson, Tomasz; Wang, Wenjian; Hodelin, Juan; Forrester, Thomas; Romanov, Volodymyr; Kostrzewski, Andrew
2016-05-01
In this paper, Bayesian Binary Sensing (BBS) is discussed as an effective tool for Bayesian Inference (BI) evaluation in interdisciplinary areas such as ISR (and, C3I), Homeland Security, QC, medicine, defense, and many others. In particular, Hilbertian Sine (HS) as an absolute measure of BI, is introduced, while avoiding relativity of decision threshold identification, as in the case of traditional measures of BI, related to false positives and false negatives.
Learning Bayesian networks for discrete data
Liang, Faming
2009-02-01
Bayesian networks have received much attention in the recent literature. In this article, we propose an approach to learn Bayesian networks using the stochastic approximation Monte Carlo (SAMC) algorithm. Our approach has two nice features. Firstly, it possesses the self-adjusting mechanism and thus avoids essentially the local-trap problem suffered by conventional MCMC simulation-based approaches in learning Bayesian networks. Secondly, it falls into the class of dynamic importance sampling algorithms; the network features can be inferred by dynamically weighted averaging the samples generated in the learning process, and the resulting estimates can have much lower variation than the single model-based estimates. The numerical results indicate that our approach can mix much faster over the space of Bayesian networks than the conventional MCMC simulation-based approaches. © 2008 Elsevier B.V. All rights reserved.
Analysis of Gumbel Model for Software Reliability Using Bayesian Paradigm
Raj Kumar
2012-12-01
Full Text Available In this paper, we have illustrated the suitability of Gumbel Model for software reliability data. The model parameters are estimated using likelihood based inferential procedure: classical as well as Bayesian. The quasi Newton-Raphson algorithm is applied to obtain the maximum likelihood estimates and associated probability intervals. The Bayesian estimates of the parameters of Gumbel model are obtained using Markov Chain Monte Carlo(MCMC simulation method in OpenBUGS(established software for Bayesian analysis using Markov Chain Monte Carlo methods. The R functions are developed to study the statistical properties, model validation and comparison tools of the model and the output analysis of MCMC samples generated from OpenBUGS. Details of applying MCMC to parameter estimation for the Gumbel model are elaborated and a real software reliability data set is considered to illustrate the methods of inference discussed in this paper.
Stochastic Collapsed Variational Bayesian Inference for Latent Dirichlet Allocation
J. Foulds; L. Boyles; C. DuBois; P. Smyth; M. Welling
2013-01-01
There has been an explosion in the amount of digital text information available in recent years, leading to challenges of scale for traditional inference algorithms for topic models. Recent advances in stochastic variational inference algorithms for latent Dirichlet allocation (LDA) have made it fea
LIU; Jianfeng; ZHANG; Yuan; ZHANG; Qin; WANG; Lixian; ZHANG; Jigang
2006-01-01
It is a challenging issue to map Quantitative Trait Loci (QTL) underlying complex discrete traits, which usually show discontinuous distribution and less information, using conventional statistical methods. Bayesian-Markov chain Monte Carlo (Bayesian-MCMC) approach is the key procedure in mapping QTL for complex binary traits, which provides a complete posterior distribution for QTL parameters using all prior information. As a consequence, Bayesian estimates of all interested variables can be obtained straightforwardly basing on their posterior samples simulated by the MCMC algorithm. In our study, utilities of Bayesian-MCMC are demonstrated using simulated several animal outbred full-sib families with different family structures for a complex binary trait underlied by both a QTL and polygene. Under the Identity-by-Descent-Based variance component random model, three samplers basing on MCMC, including Gibbs sampling, Metropolis algorithm and reversible jump MCMC, were implemented to generate the joint posterior distribution of all unknowns so that the QTL parameters were obtained by Bayesian statistical inferring. The results showed that Bayesian-MCMC approach could work well and robust under different family structures and QTL effects. As family size increases and the number of family decreases, the accuracy of the parameter estimates will be improved. When the true QTL has a small effect, using outbred population experiment design with large family size is the optimal mapping strategy.
A localization model to localize multiple sources using Bayesian inference
Dunham, Joshua Rolv
Accurate localization of a sound source in a room setting is important in both psychoacoustics and architectural acoustics. Binaural models have been proposed to explain how the brain processes and utilizes the interaural time differences (ITDs) and interaural level differences (ILDs) of sound waves arriving at the ears of a listener in determining source location. Recent work shows that applying Bayesian methods to this problem is proving fruitful. In this thesis, pink noise samples are convolved with head-related transfer functions (HRTFs) and compared to combinations of one and two anechoic speech signals convolved with different HRTFs or binaural room impulse responses (BRIRs) to simulate room positions. Through exhaustive calculation of Bayesian posterior probabilities and using a maximal likelihood approach, model selection will determine the number of sources present, and parameter estimation will result in azimuthal direction of the source(s).
Bayesian inference for a wavefront model of the Neolithisation of Europe
Baggaley, Andrew W; Shukurov, Anvar; Boys, Richard J; Golightly, Andrew
2012-01-01
We consider a wavefront model for the spread of Neolithic culture across Europe, and use Bayesian inference techniques to provide estimates for the parameters within this model, as constrained by radiocarbon data from Southern and Western Europe. Our wavefront model allows for both an isotropic background spread (incorporating the effects of local geography), and a localized anisotropic spread associated with major waterways. We introduce an innovative numerical scheme to track the wavefront, allowing us to simulate the times of the first arrival at any site orders of magnitude more efficiently than traditional PDE approaches. We adopt a Bayesian approach to inference and use Gaussian process emulators to facilitate further increases in efficiency in the inference scheme, thereby making Markov chain Monte Carlo methods practical. We allow for uncertainty in the fit of our model, and also infer a parameter specifying the magnitude of this uncertainty. We obtain a magnitude for the background spread of order 1 ...
What is the `relevant population' in Bayesian forensic inference?
Brümmer, Niko; de Villiers, Edward
2014-01-01
In works discussing the Bayesian paradigm for presenting forensic evidence in court, the concept of a `relevant population' is often mentioned, without a clear definition of what is meant, and without recommendations of how to select such populations. This note is to try to better understand this concept. Our analysis is intended to be general enough to be applicable to different forensic technologies and we shall consider both DNA profiling and speaker recognition as examples.
Robust Bayesian inference in Iq-Spherical models
Osiewalski, Jacek; Mark F.J. Steel
1992-01-01
The class of multivariate lq-spherical distributions is introduced and defined through their isodensity surfaces. We prove that, under a Jeffreys' type improper prior on the scale parameter, posterior inference on the location parameters is the same for all lq-spherical sampling models with common q. This gives us perfect inference robustness with respect to any departures from the reference case of independent sampling from the exponential power distribution.
Cornuet, Jean-Marie; Santos, Filipe; Beaumont, Mark A; Robert, Christian P.; Marin, Jean-Michel; Balding, David J.; Guillemaud, Thomas; Estoup, Arnaud
2008-01-01
Summary: Genetic data obtained on population samples convey information about their evolutionary history. Inference methods can extract part of this information but they require sophisticated statistical techniques that have been made available to the biologist community (through computer programs) only for simple and standard situations typically involving a small number of samples. We propose here a computer program (DIY ABC) for inference based on approximate Bayesian computation (ABC), in...
deBInfer: Bayesian inference for dynamical models of biological systems in R
Boersch-Supan, Philipp H.; Johnson, Leah R
2016-01-01
1. Differential equations (DEs) are commonly used to model the temporal evolution of biological systems, but statistical methods for comparing DE models to data and for parameter inference are relatively poorly developed. This is especially problematic in the context of biological systems where observations are often noisy and only a small number of time points may be available. 2. Bayesian approaches offer a coherent framework for parameter inference that can account for multiple sources of ...
Moritz eBoos
2016-05-01
Full Text Available Cognitive determinants of probabilistic inference were examined using hierarchical Bayesian modelling techniques. A classic urn-ball paradigm served as experimental strategy, involving a factorial two (prior probabilities by two (likelihoods design. Five computational models of cognitive processes were compared with the observed behaviour. Parameter-free Bayesian posterior probabilities and parameter-free base rate neglect provided inadequate models of probabilistic inference. The introduction of distorted subjective probabilities yielded more robust and generalizable results. A general class of (inverted S-shaped probability weighting functions had been proposed; however, the possibility of large differences in probability distortions not only across experimental conditions, but also across individuals, seems critical for the model’s success. It also seems advantageous to consider individual differences in parameters of probability weighting as being sampled from weakly informative prior distributions of individual parameter values. Thus, the results from hierarchical Bayesian modelling converge with previous results in revealing that probability weighting parameters show considerable task dependency and individual differences. Methodologically, this work exemplifies the usefulness of hierarchical Bayesian modelling techniques for cognitive psychology. Theoretically, human probabilistic inference might be best described as the application of individualized strategic policies for Bayesian belief revision.
Strategies for MCMC computation in quantitative genetics
Waagepetersen, Rasmus; Ibánez, N.; Sorensen, Daniel
2006-01-01
sparse inverse. Maximum likelihood inference and Bayesian inference for the linear mixed model are well-studied topics (Sorensen and Gianola, 2002). Regarding Bayesian inference, with appropriate choice of priors, the full conditional distributions are standard distributions and Gibbs sampling can be...
Perkins, Simon; Zwart, Jonathan; Natarajan, Iniyan; Smirnov, Oleg
2015-01-01
We present Montblanc, a GPU implementation of the Radio interferometer measurement equation (RIME) in support of the Bayesian inference for radio observations (BIRO) technique. BIRO uses Bayesian inference to select sky models that best match the visibilities observed by a radio interferometer. To accomplish this, BIRO evaluates the RIME multiple times, varying sky model parameters to produce multiple model visibilities. Chi-squared values computed from the model and observed visibilities are used as likelihood values to drive the Bayesian sampling process and select the best sky model. As most of the elements of the RIME and chi-squared calculation are independent of one another, they are highly amenable to parallel computation. Additionally, Montblanc caters for iterative RIME evaluation to produce multiple chi-squared values. Only modified model parameters are transferred to the GPU between each iteration. We implemented Montblanc as a Python package based upon NVIDIA's CUDA architecture. As such, it is ea...
Probabilistic Modelling of Fatigue Life of Composite Laminates Using Bayesian Inference
Dimitrov, Nikolay Krasimirov; Kiureghian, Armen Der
2014-01-01
. Model parameters are estimated by Bayesian inference. The reference data used consists of constant-amplitude fatigue test results for a multi-directional laminate subjected to seven different load ratios. The paper describes the modelling techniques and the parameter estimation procedure, supported by...
Markov Model of Wind Power Time Series UsingBayesian Inference of Transition Matrix
Chen, Peiyuan; Berthelsen, Kasper Klitgaard; Bak-Jensen, Birgitte;
2009-01-01
This paper proposes to use Bayesian inference of transition matrix when developing a discrete Markov model of a wind speed/power time series and 95% credible interval for the model verification. The Dirichlet distribution is used as a conjugate prior for the transition matrix. Three discrete Markov...
Internal dosimetry of uranium isotopes using bayesian inference methods
A group of personnel at Los Alamos National Laboratory is routinely monitored for the presence of uranium isotopes by urine bioassay. Samples are analysed by alpha spectroscopy, and the results are examined for evidence of an intake of uranium. Because the measurement uncertainties are often comparable to the quantities of material we wish to detect, statistical considerations are crucial for the proper interpretation of the data. The problem is further complicated by the significant, but highly non-uniform, presence of uranium in local drinking water and, in some cases, food supply. Software originally developed for internal dosimetry of plutonium has been adapted to the problem of uranium dosimetry. The software uses an unfolding algorithm to calculate an approximate Bayesian solution to the problem of characterising any intakes which may have occurred, given the history of urine bioassay results for each individual in the monitored population. The program uses biokinetic models from ICRP Publications 68 and later, and a prior probability distribution derived empirically from the body of uranium bioassay data collected at Los Alamos over the operating history of the Laboratory. For each individual, the software creates a posterior probability distribution of intake quantity and solubility type as a function of time. From this distribution, estimates are made of the cumulative committed dose (CEDE) to each individual. Results of the method are compared with those obtained using an earlier classical (non-Bayesian) algorithm for uranium dosimetry. We also discuss the problem of distinguishing occupational intakes from intake of environmental uranium, within a Bayesian framework. (author)
Using SAS PROC MCMC for Item Response Theory Models
Ames, Allison J.; Samonte, Kelli
2015-01-01
Interest in using Bayesian methods for estimating item response theory models has grown at a remarkable rate in recent years. This attentiveness to Bayesian estimation has also inspired a growth in available software such as WinBUGS, R packages, BMIRT, MPLUS, and SAS PROC MCMC. This article intends to provide an accessible overview of Bayesian…
Matrix-free Large Scale Bayesian inference in cosmology
Jasche, Jens
2014-01-01
In this work we propose a new matrix-free implementation of the Wiener sampler which is traditionally applied to high dimensional analysis when signal covariances are unknown. Specifically, the proposed method addresses the problem of jointly inferring a high dimensional signal and its corresponding covariance matrix from a set of observations. Our method implements a Gibbs sampling adaptation of the previously presented messenger approach, permitting to cast the complex multivariate inference problem into a sequence of uni-variate random processes. In this fashion, the traditional requirement of inverting high dimensional matrices is completely eliminated from the inference process, resulting in an efficient algorithm that is trivial to implement. Using cosmic large scale structure data as a showcase, we demonstrate the capabilities of our Gibbs sampling approach by performing a joint analysis of three dimensional density fields and corresponding power-spectra from Gaussian mock catalogues. These tests clear...
Bayesian Inference on Predictors of Sex of the Baby
Scarpa, Bruno
2016-01-01
It is well known that the sex ratio at birth is a biological constant, being about 106 boys to 100 girls. However couples have always wanted to know and decide in advance the sex of a newborn. For example, a large number of papers appeared connecting biometrical variables, such as length of follicular phase in the woman menstrual cycle or timing of intercourse acts to the sex of new baby. In this paper, we propose a Bayesian model to validate some of these theories by using an independent dat...
Bayesian inference and model comparison for metallic fatigue data
Babuška, Ivo
2016-02-23
In this work, we present a statistical treatment of stress-life (S-N) data drawn from a collection of records of fatigue experiments that were performed on 75S-T6 aluminum alloys. Our main objective is to predict the fatigue life of materials by providing a systematic approach to model calibration, model selection and model ranking with reference to S-N data. To this purpose, we consider fatigue-limit models and random fatigue-limit models that are specially designed to allow the treatment of the run-outs (right-censored data). We first fit the models to the data by maximum likelihood methods and estimate the quantiles of the life distribution of the alloy specimen. To assess the robustness of the estimation of the quantile functions, we obtain bootstrap confidence bands by stratified resampling with respect to the cycle ratio. We then compare and rank the models by classical measures of fit based on information criteria. We also consider a Bayesian approach that provides, under the prior distribution of the model parameters selected by the user, their simulation-based posterior distributions. We implement and apply Bayesian model comparison methods, such as Bayes factor ranking and predictive information criteria based on cross-validation techniques under various a priori scenarios.
Bayesian inference and model comparison for metallic fatigue data
Babuška, Ivo; Sawlan, Zaid; Scavino, Marco; Szabó, Barna; Tempone, Raúl
2016-06-01
In this work, we present a statistical treatment of stress-life (S-N) data drawn from a collection of records of fatigue experiments that were performed on 75S-T6 aluminum alloys. Our main objective is to predict the fatigue life of materials by providing a systematic approach to model calibration, model selection and model ranking with reference to S-N data. To this purpose, we consider fatigue-limit models and random fatigue-limit models that are specially designed to allow the treatment of the run-outs (right-censored data). We first fit the models to the data by maximum likelihood methods and estimate the quantiles of the life distribution of the alloy specimen. To assess the robustness of the estimation of the quantile functions, we obtain bootstrap confidence bands by stratified resampling with respect to the cycle ratio. We then compare and rank the models by classical measures of fit based on information criteria. We also consider a Bayesian approach that provides, under the prior distribution of the model parameters selected by the user, their simulation-based posterior distributions. We implement and apply Bayesian model comparison methods, such as Bayes factor ranking and predictive information criteria based on cross-validation techniques under various a priori scenarios.
Hierarchical Bayesian inference of galaxy redshift distributions from photometric surveys
Leistedt, Boris; Peiris, Hiranya V
2016-01-01
Accurately characterizing the redshift distributions of galaxies is essential for analysing deep photometric surveys and testing cosmological models. We present a technique to simultaneously infer redshift distributions and individual redshifts from photometric galaxy catalogues. Our model constructs a piecewise constant representation (effectively a histogram) of the distribution of galaxy types and redshifts, the parameters of which are efficiently inferred from noisy photometric flux measurements. This approach can be seen as a generalization of template-fitting photometric redshift methods and relies on a library of spectral templates to relate the photometric fluxes of individual galaxies to their redshifts. We illustrate this technique on simulated galaxy survey data, and demonstrate that it delivers correct posterior distributions on the underlying type and redshift distributions, as well as on the individual types and redshifts of galaxies. We show that even with uninformative priors, large photometri...
Approximate Bayesian inference in semi-mechanistic models
Aderhold, Andrej; Husmeier, Dirk; Grzegorczyk, Marco
2016-01-01
Inference of interaction networks represented by systems of differential equations is a challenging problem in many scientific disciplines. In the present article, we follow a semi-mechanistic modelling approach based on gradient matching. We investigate the extent to which key factors, including the kinetic model, statistical formulation and numerical methods, impact upon performance at network reconstruction. We emphasize general lessons for computational statisticians when faced with the c...
Solving #SAT and Bayesian Inference with Backtracking Search
Bacchus, Fahiem; Dalmao, Shannon; Pitassi, Toniann
2014-01-01
Inference in Bayes Nets (BAYES) is an important problem with numerous applications in probabilistic reasoning. Counting the number of satisfying assignments of a propositional formula (#SAT) is a closely related problem of fundamental theoretical importance. Both these problems, and others, are members of the class of sum-of-products (SUMPROD) problems. In this paper we show that standard backtracking search when augmented with a simple memoization scheme (caching) can solve any sum-of-produc...
Wu, Yuefeng; Hooker, Giles
2013-01-01
This paper introduces a hierarchical framework to incorporate Hellinger distance methods into Bayesian analysis. We propose to modify a prior over non-parametric densities with the exponential of twice the Hellinger distance between a candidate and a parametric density. By incorporating a prior over the parameters of the second density, we arrive at a hierarchical model in which a non-parametric model is placed between parameters and the data. The parameters of the family can then be estimate...
Hierarchical Bayesian inference of galaxy redshift distributions from photometric surveys
Leistedt, Boris; Mortlock, Daniel J.; Peiris, Hiranya V.
2016-08-01
Accurately characterizing the redshift distributions of galaxies is essential for analysing deep photometric surveys and testing cosmological models. We present a technique to simultaneously infer redshift distributions and individual redshifts from photometric galaxy catalogues. Our model constructs a piecewise constant representation (effectively a histogram) of the distribution of galaxy types and redshifts, the parameters of which are efficiently inferred from noisy photometric flux measurements. This approach can be seen as a generalization of template-fitting photometric redshift methods and relies on a library of spectral templates to relate the photometric fluxes of individual galaxies to their redshifts. We illustrate this technique on simulated galaxy survey data, and demonstrate that it delivers correct posterior distributions on the underlying type and redshift distributions, as well as on the individual types and redshifts of galaxies. We show that even with uninformative priors, large photometric errors and parameter degeneracies, the redshift and type distributions can be recovered robustly thanks to the hierarchical nature of the model, which is not possible with common photometric redshift estimation techniques. As a result, redshift uncertainties can be fully propagated in cosmological analyses for the first time, fulfilling an essential requirement for the current and future generations of surveys.
Large-Scale Optimization for Bayesian Inference in Complex Systems
Willcox, Karen [MIT; Marzouk, Youssef [MIT
2013-11-12
The SAGUARO (Scalable Algorithms for Groundwater Uncertainty Analysis and Robust Optimization) Project focused on the development of scalable numerical algorithms for large-scale Bayesian inversion in complex systems that capitalize on advances in large-scale simulation-based optimization and inversion methods. The project was a collaborative effort among MIT, the University of Texas at Austin, Georgia Institute of Technology, and Sandia National Laboratories. The research was directed in three complementary areas: efficient approximations of the Hessian operator, reductions in complexity of forward simulations via stochastic spectral approximations and model reduction, and employing large-scale optimization concepts to accelerate sampling. The MIT--Sandia component of the SAGUARO Project addressed the intractability of conventional sampling methods for large-scale statistical inverse problems by devising reduced-order models that are faithful to the full-order model over a wide range of parameter values; sampling then employs the reduced model rather than the full model, resulting in very large computational savings. Results indicate little effect on the computed posterior distribution. On the other hand, in the Texas--Georgia Tech component of the project, we retain the full-order model, but exploit inverse problem structure (adjoint-based gradients and partial Hessian information of the parameter-to-observation map) to implicitly extract lower dimensional information on the posterior distribution; this greatly speeds up sampling methods, so that fewer sampling points are needed. We can think of these two approaches as ``reduce then sample'' and ``sample then reduce.'' In fact, these two approaches are complementary, and can be used in conjunction with each other. Moreover, they both exploit deterministic inverse problem structure, in the form of adjoint-based gradient and Hessian information of the underlying parameter-to-observation map, to
Bayesian Spatial Modelling with R-INLA
Finn Lindgren; Håvard Rue
2015-01-01
The principles behind the interface to continuous domain spatial models in the R- INLA software package for R are described. The integrated nested Laplace approximation (INLA) approach proposed by Rue, Martino, and Chopin (2009) is a computationally effective alternative to MCMC for Bayesian inference. INLA is designed for latent Gaussian models, a very wide and flexible class of models ranging from (generalized) linear mixed to spatial and spatio-temporal models. Combined with the stochastic...
Dorn, C; Khan, A; Heng, K; Alibert, Y; Helled, R; Rivoldini, A; Benz, W
2016-01-01
We aim to present a generalized Bayesian inference method for constraining interiors of super Earths and sub-Neptunes. Our methodology succeeds in quantifying the degeneracy and correlation of structural parameters for high dimensional parameter spaces. Specifically, we identify what constraints can be placed on composition and thickness of core, mantle, ice, ocean, and atmospheric layers given observations of mass, radius, and bulk refractory abundance constraints (Fe, Mg, Si) from observations of the host star's photospheric composition. We employed a full probabilistic Bayesian inference analysis that formally accounts for observational and model uncertainties. Using a Markov chain Monte Carlo technique, we computed joint and marginal posterior probability distributions for all structural parameters of interest. We included state-of-the-art structural models based on self-consistent thermodynamics of core, mantle, high-pressure ice, and liquid water. Furthermore, we tested and compared two different atmosp...
Ball, William T; Egerton, Jack S; Haigh, Joanna D
2014-01-01
We investigate the relationship between spectral solar irradiance (SSI) and ozone in the tropical upper stratosphere. We find that solar cycle (SC) changes in ozone can be well approximated by considering the ozone response to SSI changes in a small number individual wavelength bands between 176 and 310 nm, operating independently of each other. Additionally, we find that the ozone varies approximately linearly with changes in the SSI. Using these facts, we present a Bayesian formalism for inferring SC SSI changes and uncertainties from measured SC ozone profiles. Bayesian inference is a powerful, mathematically self-consistent method of considering both the uncertainties of the data and additional external information to provide the best estimate of parameters being estimated. Using this method, we show that, given measurement uncertainties in both ozone and SSI datasets, it is not currently possible to distinguish between observed or modelled SSI datasets using available estimates of ozone change profiles, ...
Prudhomme, Serge
2015-09-17
Parameter estimation for complex models using Bayesian inference is usually a very costly process as it requires a large number of solves of the forward problem. We show here how the construction of adaptive surrogate models using a posteriori error estimates for quantities of interest can significantly reduce the computational cost in problems of statistical inference. As surrogate models provide only approximations of the true solutions of the forward problem, it is nevertheless necessary to control these errors in order to construct an accurate reduced model with respect to the observables utilized in the identification of the model parameters. Effectiveness of the proposed approach is demonstrated on a numerical example dealing with the Spalart–Allmaras model for the simulation of turbulent channel flows. In particular, we illustrate how Bayesian model selection using the adapted surrogate model in place of solving the coupled nonlinear equations leads to the same quality of results while requiring fewer nonlinear PDE solves.
Dorazio, R.M.; Johnson, F.A.
2003-01-01
Bayesian inference and decision theory may be used in the solution of relatively complex problems of natural resource management, owing to recent advances in statistical theory and computing. In particular, Markov chain Monte Carlo algorithms provide a computational framework for fitting models of adequate complexity and for evaluating the expected consequences of alternative management actions. We illustrate these features using an example based on management of waterfowl habitat.
Bayesian inference of models and hyper-parameters for robust optic-flow estimation
Héas, Patrick; Herzet, Cédric; Memin, Etienne
2012-01-01
International audience Selecting optimal models and hyper-parameters is crucial for accurate optic-flow estimation. This paper provides a solution to the problem in a generic Bayesian framework. The method is based on a conditional model linking the image intensity function, the unknown velocity field, hyper-parameters and the prior and likelihood motion models. Inference is performed on each of the three-level of this so-defined hierarchical model by maximization of marginalized \\textit{a...
GNU MCSim : bayesian statistical inference for SBML-coded systems biology models
Bois, Frédéric Y.
2009-01-01
International audience Statistical inference about the parameter values of complex models, such as the ones routinely developed in systems biology, is efficiently performed through Bayesian numerical techniques. In that framework, prior information and multiple levels of uncertainty can be seamlessly integrated. GNU MCSim was precisely developed to achieve those aims, in a general non-linear differential context. Starting with version 5.3.0, GNU MCSim reads in and simulates Systems Biology...
Bayesian inferences of the thermal properties of a wall using temperature and heat flux measurements
Iglesias, Marco; Sawlan, Zaid; Scavino, Marco; Tempone, Raul; Wood, Christopher
2016-01-01
We develop a hierarchical Bayesian inference method to estimate the thermal resistance and the volumetric heat capacity of a wall. These thermal properties are essential for accurate building energy simulations that are needed to make effective energy-saving policies. We apply our methodology to an experimental case study conducted in an environmental chamber, where measurements are recorded every minute from temperature probes and heat flux sensors placed on both sides of a solid brick wall ...
Seyed Taghi Akhavan Niaki; Mohammad Saber Fallah Nezhad
2007-01-01
In order to design a decision-making framework in production environments, in this study, we use both the stochastic dynamic programming and Bayesian inference concepts. Using the posterior probability of the production process to be in state λ (the hazard rate of defective products), first we formulate the problem into a stochastic dynamic programming model. Next, we derive some properties for the optimal value of the objective function. Then, we propose a solution algorithm. At the end, the...
Constraining East Antarctic mass trends using a Bayesian inference approach
Martin-Español, Alba; Bamber, Jonathan L.
2016-04-01
East Antarctica is an order of magnitude larger than its western neighbour and the Greenland ice sheet. It has the greatest potential to contribute to sea level rise of any source, including non-glacial contributors. It is, however, the most challenging ice mass to constrain because of a range of factors including the relative paucity of in-situ observations and the poor signal to noise ratio of Earth Observation data such as satellite altimetry and gravimetry. A recent study using satellite radar and laser altimetry (Zwally et al. 2015) concluded that the East Antarctic Ice Sheet (EAIS) had been accumulating mass at a rate of 136±28 Gt/yr for the period 2003-08. Here, we use a Bayesian hierarchical model, which has been tested on, and applied to, the whole of Antarctica, to investigate the impact of different assumptions regarding the origin of elevation changes of the EAIS. We combined GRACE, satellite laser and radar altimeter data and GPS measurements to solve simultaneously for surface processes (primarily surface mass balance, SMB), ice dynamics and glacio-isostatic adjustment over the period 2003-13. The hierarchical model partitions mass trends between SMB and ice dynamics based on physical principles and measures of statistical likelihood. Without imposing the division between these processes, the model apportions about a third of the mass trend to ice dynamics, +18 Gt/yr, and two thirds, +39 Gt/yr, to SMB. The total mass trend for that period for the EAIS was 57±20 Gt/yr. Over the period 2003-08, we obtain an ice dynamic trend of 12 Gt/yr and a SMB trend of 15 Gt/yr, with a total mass trend of 27 Gt/yr. We then imposed the condition that the surface mass balance is tightly constrained by the regional climate model RACMO2.3 and allowed height changes due to ice dynamics to occur in areas of low surface velocities (<10 m/yr) , such as those in the interior of East Antarctica (a similar condition as used in Zwally 2015). The model must find a solution that
Adaptability and phenotypic stability of common bean genotypes through Bayesian inference.
Corrêa, A M; Teodoro, P E; Gonçalves, M C; Barroso, L M A; Nascimento, M; Santos, A; Torres, F E
2016-01-01
This study used Bayesian inference to investigate the genotype x environment interaction in common bean grown in Mato Grosso do Sul State, and it also evaluated the efficiency of using informative and minimally informative a priori distributions. Six trials were conducted in randomized blocks, and the grain yield of 13 common bean genotypes was assessed. To represent the minimally informative a priori distributions, a probability distribution with high variance was used, and a meta-analysis concept was adopted to represent the informative a priori distributions. Bayes factors were used to conduct comparisons between the a priori distributions. The Bayesian inference was effective for the selection of upright common bean genotypes with high adaptability and phenotypic stability using the Eberhart and Russell method. Bayes factors indicated that the use of informative a priori distributions provided more accurate results than minimally informative a priori distributions. According to Bayesian inference, the EMGOPA-201, BAMBUÍ, CNF 4999, CNF 4129 A 54, and CNFv 8025 genotypes had specific adaptability to favorable environments, while the IAPAR 14 and IAC CARIOCA ETE genotypes had specific adaptability to unfavorable environments. PMID:27173270
Kang, Seong Keun; Seong, Poong Hyun [KAIST, Daejeon (Korea, Republic of)
2014-08-15
Bayesian methodology has been used widely used in various research fields. It is method of inference using Bayes' rule to update the estimation of probability for the certain hypothesis when additional evidences are acquired. According to the current researches, malfunction of nuclear power plant can be detected by using this Bayesian inference which consistently piles up the newly incoming data and updates its estimation. However, those researches are based on the assumption that people are doing like computer perfectly, which can be criticized and may cause a problem in real world application. Studies in cognitive psychology indicates that when the amount of information becomes larger, people can't save the whole data because people have limited memory capacity which is well known as working memory, and also they have attention problem. The purpose of this paper is to consider the psychological factors and confirm how much this working memory and attention will affect the resulted estimation based on the Bayesian inference. To confirm this, experiment on human is needed, and the tool of experiment is Compact Nuclear Simulator (CNS)
Bayesian methodology has been used widely used in various research fields. It is method of inference using Bayes' rule to update the estimation of probability for the certain hypothesis when additional evidences are acquired. According to the current researches, malfunction of nuclear power plant can be detected by using this Bayesian inference which consistently piles up the newly incoming data and updates its estimation. However, those researches are based on the assumption that people are doing like computer perfectly, which can be criticized and may cause a problem in real world application. Studies in cognitive psychology indicates that when the amount of information becomes larger, people can't save the whole data because people have limited memory capacity which is well known as working memory, and also they have attention problem. The purpose of this paper is to consider the psychological factors and confirm how much this working memory and attention will affect the resulted estimation based on the Bayesian inference. To confirm this, experiment on human is needed, and the tool of experiment is Compact Nuclear Simulator (CNS)
A novel multimode process monitoring method integrating LDRSKM with Bayesian inference
Shi-jin REN; Yin LIANG; Xiang-jun ZHAO; Mao-yun YANG
2015-01-01
A local discriminant regularized soft k-means (LDRSKM) method with Bayesian inference is proposed for multimode process monitoring. LDRSKM extends the regularized soft k-means algorithm by exploiting the local and non-local geometric information of the data and generalized linear discriminant analysis to provide a better and more meaningful data partition. LDRSKM can perform clustering and subspace selection simultaneously, enhancing the separability of data residing in different clusters. With the data partition obtained, kernel support vector data description (KSVDD) is used to establish the monitoring statistics and control limits. Two Bayesian inference based global fault detection indicators are then developed using the local monitoring results associated with principal and residual subspaces. Based on clustering analysis, Bayesian inference and mani-fold learning methods, the within and cross-mode correlations, and local geometric information can be exploited to enhance monitoring performances for nonlinear and non-Gaussian processes. The effectiveness and efficiency of the proposed method are evaluated using the Tennessee Eastman benchmark process.
Trans-dimensional Bayesian inference for large sequential data sets
Mandolesi, E.; Dettmer, J.; Dosso, S. E.; Holland, C. W.
2015-12-01
This work develops a sequential Monte Carlo method to infer seismic parameters of layered seabeds from large sequential reflection-coefficient data sets. The approach provides parameter estimates and uncertainties along survey tracks with the goal to aid in the detection of unexploded ordnance in shallow water. The sequential data are acquired by a moving platform with source and receiver array towed close to the seabed. This geometry requires consideration of spherical reflection coefficients, computed efficiently by massively parallel implementation of the Sommerfeld integral via Levin integration on a graphics processing unit. The seabed is parametrized with a trans-dimensional model to account for changes in the environment (i.e. changes in layering) along the track. The method combines advanced Markov chain Monte Carlo methods (annealing) with particle filtering (resampling). Since data from closely-spaced source transmissions (pings) often sample similar environments, the solution from one ping can be utilized to efficiently estimate the posterior for data from subsequent pings. Since reflection-coefficient data are highly informative, the likelihood function can be extremely peaked, resulting in little overlap between posteriors of adjacent pings. This is addressed by adding bridging distributions (via annealed importance sampling) between pings for more efficient transitions. The approach assumes the environment to be changing slowly enough to justify the local 1D parametrization. However, bridging allows rapid changes between pings to be addressed and we demonstrate the method to be stable in such situations. Results are in terms of trans-D parameter estimates and uncertainties along the track. The algorithm is examined for realistic simulated data along a track and applied to a dataset collected by an autonomous underwater vehicle on the Malta Plateau, Mediterranean Sea. [Work supported by the SERDP, DoD.
Bayesian inference of T Tauri star properties using multi-wavelength survey photometry
Barentsen, Geert; Drew, Janet E; Sale, Stuart E
2012-01-01
There are many pertinent open issues in the area of star and planet formation. Large statistical samples of young stars across star-forming regions are needed to trigger a breakthrough in our understanding, but most optical studies are based on a wide variety of spectrographs and analysis methods, which introduces large biases. Here we show how graphical Bayesian networks can be employed to construct a hierarchical probabilistic model which allows pre-main sequence ages, masses, accretion rates, and extinctions to be estimated using two widely available photometric survey databases (IPHAS r/i/Halpha and 2MASS J-band magnitudes.) Because our approach does not rely on spectroscopy, it can easily be applied to homogeneously study the large number of clusters for which Gaia will yield membership lists. We explain how the analysis is carried out using the Markov Chain Monte Carlo (MCMC) method and provide Python source code. We then demonstrate its use on 587 known low-mass members of the star-forming region NGC 2...
Inferring activities from context measurements using Bayesian inference and random utility models
Hurtubia, Ricardo; Bierlaire, Michel; Flötteröd, Gunnar
2009-01-01
Smartphones collect a wealth of information about their users. This includes GPS tracks and the MAC addresses of devices around the user, and it can go as far as taking visual and acoustic samples of the user's environment. We present a framework to identify a smartphone user's activities in a Bayesian setting. As prior information, we us a random utility model that accounts for the type of activity a user is likely to perform at any given location and time; this model was estimated for the w...
Marcelo Costa Souza
2004-10-01
, in which the parameters are regarded as fixed quantities, not assuming changes in time. This work aimed at fitting of autoregressive models with order 2, AR(2, specified in the form of dynamic linear models using Bayesian inference. Monte Carlo Markov Chain (MCMC was used to obtain the estimates, via Gibbs Sampler and Forward Filtering Backward Sampling (FFBS. To evaluate the fitting, two chains with 8000 iterations each, and three different series sizes, with 200, 500 and 800 observations were sampled. The Canadian lynx series (NICHOLLS and QUIN, 1982, was fitted with different discount factors (0.90, 0.95 and 0.99, and the resulting mean square error was used to compare to the fitting using classical inference. A better fit for the model with discount equal to 0.99 was observed. One-step ahead forecasts were done to check the estimates obtained for the updated and the backward sampled series. To the latter, the fitting was better and mean square error lower. In general, it was observed a good fit of the AR(2 dynamic models via Bayesian inference, and this gives a better understanding of the fitting in different situations, both simulated and real.
Feroz, Farhan
2007-01-01
In performing a Bayesian analysis of astronomical data, two difficult problems often emerge. First, in estimating the parameters of some model for the data, the resulting posterior distribution may be multimodal or exhibit pronounced (curving) degeneracies, which can cause problems for traditional MCMC sampling methods. Second, in selecting between a set of competing models, calculation of the Bayesian evidence for each model is computationally expensive. The nested sampling method introduced by Skilling (2004), has greatly reduced the computational expense of calculating evidences and also produces posterior inferences as a by-product. This method has been applied successfully in cosmological applications by Mukherjee et al. (2006), but their implementation was efficient only for unimodal distributions without pronounced degeneracies. Shaw et al. (2007), recently introduced a clustered nested sampling method which is significantly more efficient in sampling from multimodal posteriors and also determines the ...
AGNfitter: SED-fitting code for AGN and galaxies from a MCMC approach
Calistro Rivera, Gabriela; Lusso, Elisabeta; Hennawi, Joseph F.; Hogg, David W.
2016-07-01
AGNfitter is a fully Bayesian MCMC method to fit the spectral energy distributions (SEDs) of active galactic nuclei (AGN) and galaxies from the sub-mm to the UV; it enables robust disentanglement of the physical processes responsible for the emission of sources. Written in Python, AGNfitter makes use of a large library of theoretical, empirical, and semi-empirical models to characterize both the nuclear and host galaxy emission simultaneously. The model consists of four physical emission components: an accretion disk, a torus of AGN heated dust, stellar populations, and cold dust in star forming regions. AGNfitter determines the posterior distributions of numerous parameters that govern the physics of AGN with a fully Bayesian treatment of errors and parameter degeneracies, allowing one to infer integrated luminosities, dust attenuation parameters, stellar masses, and star formation rates.
Gaffney, Jim A; Sonnad, Vijay; Libby, Stephen B
2013-01-01
First principles microphysics models are essential to the design and analysis of high energy density physics experiments. Using experimental data to investigate the underlying physics is also essential, particularly when simulations and experiments are not consistent with each other. This is a difficult task, due to the large number of physical models that play a role, and due to the complex (and as a result, noisy) nature of the experiments. This results in a large number of parameters that make any inference a daunting task; it is also very important to consistently treat both experimental and prior understanding of the problem. In this paper we present a Bayesian method that includes both these effects, and allows the inference of a set of modifiers which have been constructed to give information about microphysics models from experimental data. We pay particular attention to radiation transport models. The inference takes into account a large set of experimental parameters and an estimate of the prior kno...
Uncertainty Analysis in Fatigue Life Prediction of Gas Turbine Blades Using Bayesian Inference
Li, Yan-Feng; Zhu, Shun-Peng; Li, Jing; Peng, Weiwen; Huang, Hong-Zhong
2015-12-01
This paper investigates Bayesian model selection for fatigue life estimation of gas turbine blades considering model uncertainty and parameter uncertainty. Fatigue life estimation of gas turbine blades is a critical issue for the operation and health management of modern aircraft engines. Since lots of life prediction models have been presented to predict the fatigue life of gas turbine blades, model uncertainty and model selection among these models have consequently become an important issue in the lifecycle management of turbine blades. In this paper, fatigue life estimation is carried out by considering model uncertainty and parameter uncertainty simultaneously. It is formulated as the joint posterior distribution of a fatigue life prediction model and its model parameters using Bayesian inference method. Bayes factor is incorporated to implement the model selection with the quantified model uncertainty. Markov Chain Monte Carlo method is used to facilitate the calculation. A pictorial framework and a step-by-step procedure of the Bayesian inference method for fatigue life estimation considering model uncertainty are presented. Fatigue life estimation of a gas turbine blade is implemented to demonstrate the proposed method.
Planetary micro-rover operations on Mars using a Bayesian framework for inference and control
Post, Mark A.; Li, Junquan; Quine, Brendan M.
2016-03-01
With the recent progress toward the application of commercially-available hardware to small-scale space missions, it is now becoming feasible for groups of small, efficient robots based on low-power embedded hardware to perform simple tasks on other planets in the place of large-scale, heavy and expensive robots. In this paper, we describe design and programming of the Beaver micro-rover developed for Northern Light, a Canadian initiative to send a small lander and rover to Mars to study the Martian surface and subsurface. For a small, hardware-limited rover to handle an uncertain and mostly unknown environment without constant management by human operators, we use a Bayesian network of discrete random variables as an abstraction of expert knowledge about the rover and its environment, and inference operations for control. A framework for efficient construction and inference into a Bayesian network using only the C language and fixed-point mathematics on embedded hardware has been developed for the Beaver to make intelligent decisions with minimal sensor data. We study the performance of the Beaver as it probabilistically maps a simple outdoor environment with sensor models that include uncertainty. Results indicate that the Beaver and other small and simple robotic platforms can make use of a Bayesian network to make intelligent decisions in uncertain planetary environments.
Michael J McGeachie
2014-06-01
Full Text Available Bayesian Networks (BN have been a popular predictive modeling formalism in bioinformatics, but their application in modern genomics has been slowed by an inability to cleanly handle domains with mixed discrete and continuous variables. Existing free BN software packages either discretize continuous variables, which can lead to information loss, or do not include inference routines, which makes prediction with the BN impossible. We present CGBayesNets, a BN package focused around prediction of a clinical phenotype from mixed discrete and continuous variables, which fills these gaps. CGBayesNets implements Bayesian likelihood and inference algorithms for the conditional Gaussian Bayesian network (CGBNs formalism, one appropriate for predicting an outcome of interest from, e.g., multimodal genomic data. We provide four different network learning algorithms, each making a different tradeoff between computational cost and network likelihood. CGBayesNets provides a full suite of functions for model exploration and verification, including cross validation, bootstrapping, and AUC manipulation. We highlight several results obtained previously with CGBayesNets, including predictive models of wood properties from tree genomics, leukemia subtype classification from mixed genomic data, and robust prediction of intensive care unit mortality outcomes from metabolomic profiles. We also provide detailed example analysis on public metabolomic and gene expression datasets. CGBayesNets is implemented in MATLAB and available as MATLAB source code, under an Open Source license and anonymous download at http://www.cgbayesnets.com.
Bayesian inference of solar and stellar magnetic fields in the weak-field approximation
Ramos, A Asensio
2011-01-01
The weak-field approximation is one of the simplest models that allows us to relate the observed polarization induced by the Zeeman effect with the magnetic field vector present on the plasma of interest. It is usually applied for diagnosing magnetic fields in the solar and stellar atmospheres. A fully Bayesian approach to the inference of magnetic properties in unresolved structures is presented. The analytical expression for the marginal posterior distribution is obtained, from which we can obtain statistically relevant information about the model parameters. The role of a-priori information is discussed and a hierarchical procedure is presented that gives robust results that are almost insensitive to the precise election of the prior. The strength of the formalism is demonstrated through an application to IMaX data. Bayesian methods can optimally exploit data from filter-polarimeters given the scarcity of spectral information as compared with spectro-polarimeters. The effect of noise and how it degrades ou...
Holzinger, M.
2014-09-01
This paper introduces and discusses a method to rigorously classify and prioritize UCTs using Bayesian inference and admissible regions. A detailed derivation and discussion of the methodology is given, followed by a generalized definition of prioritization parameters. Several example prioritization parameters including `time left to detect,' 'zero-effort miss,' and `effective albedo-area' are motivated and given. A number of illustrative applications with optical UCTs are examined to demonstrate information that can be extracted from each observation. Finally, the information extracted from each UCT is then compared and approaches to observation prioritization discussed.
DIP -- Diagnostics for Insufficiencies of Posterior calculations in Bayesian signal inference
Dorn, Sebastian; lin, Torsten A Enß
2013-01-01
We present an error-diagnostic validation method for posterior distributions in Bayesian signal inference. It transfers deviations from the correct posterior into characteristic deviations from a uniform distribution of a quantity constructed for this purpose. We show that this method is able to reveal and discriminate several kinds of numerical and approximation errors. For this we present a number of analytical examples of posteriors with incorrect variance, skewness, position of the maximum, or normalization. We show further how this test can be applied to multidimensional signals.
Hadwin, Paul J.; Sipkens, T. A.; Thomson, K. A.; Liu, F.; Daun, K. J.
2016-01-01
Auto-correlated laser-induced incandescence (AC-LII) infers the soot volume fraction (SVF) of soot particles by comparing the spectral incandescence from laser-energized particles to the pyrometrically inferred peak soot temperature. This calculation requires detailed knowledge of model parameters such as the absorption function of soot, which may vary with combustion chemistry, soot age, and the internal structure of the soot. This work presents a Bayesian methodology to quantify such uncertainties. This technique treats the additional "nuisance" model parameters, including the soot absorption function, as stochastic variables and incorporates the current state of knowledge of these parameters into the inference process through maximum entropy priors. While standard AC-LII analysis provides a point estimate of the SVF, Bayesian techniques infer the posterior probability density, which will allow scientists and engineers to better assess the reliability of AC-LII inferred SVFs in the context of environmental regulations and competing diagnostics.
Occurrence of hazardous accident in nuclear power plants and industrial units usually lead to release of radioactive materials and pollutants in environment. These materials and pollutants can be transported to a far downstream by the wind flow. In this paper, we implemented an atmospheric dispersion code to solve the inverse problem. Having received and detected the pollutants in one region, we may estimate the rate and location of the unknown source. For the modeling, one needs a model with ability of atmospheric dispersion calculation. Furthermore, it is required to implement a mathematical approach to infer the source location and the related rates. In this paper the AERMOD software and Bayesian inference along the Markov Chain Monte Carlo have been applied. Implementing, Bayesian approach and Markov Chain Monte Carlo for the aforementioned subject is not a new approach, but the AERMOD model coupled with the said methods is a new and well known regulatory software, and enhances the reliability of outcomes. To evaluate the method, an example is considered by defining pollutants concentration in a specific region and then obtaining the source location and intensity by a direct calculation. The result of the calculation estimates the average source location at a distance of 7km with an accuracy of 5m which is good enough to support the ability of the proposed algorithm.
Sraj, Ihab
2015-10-22
This paper addresses model dimensionality reduction for Bayesian inference based on prior Gaussian fields with uncertainty in the covariance function hyper-parameters. The dimensionality reduction is traditionally achieved using the Karhunen-Loève expansion of a prior Gaussian process assuming covariance function with fixed hyper-parameters, despite the fact that these are uncertain in nature. The posterior distribution of the Karhunen-Loève coordinates is then inferred using available observations. The resulting inferred field is therefore dependent on the assumed hyper-parameters. Here, we seek to efficiently estimate both the field and covariance hyper-parameters using Bayesian inference. To this end, a generalized Karhunen-Loève expansion is derived using a coordinate transformation to account for the dependence with respect to the covariance hyper-parameters. Polynomial Chaos expansions are employed for the acceleration of the Bayesian inference using similar coordinate transformations, enabling us to avoid expanding explicitly the solution dependence on the uncertain hyper-parameters. We demonstrate the feasibility of the proposed method on a transient diffusion equation by inferring spatially-varying log-diffusivity fields from noisy data. The inferred profiles were found closer to the true profiles when including the hyper-parameters’ uncertainty in the inference formulation.
Perkins, S. J.; Marais, P. C.; Zwart, J. T. L.; Natarajan, I.; Tasse, C.; Smirnov, O.
2015-09-01
We present Montblanc, a GPU implementation of the Radio interferometer measurement equation (RIME) in support of the Bayesian inference for radio observations (BIRO) technique. BIRO uses Bayesian inference to select sky models that best match the visibilities observed by a radio interferometer. To accomplish this, BIRO evaluates the RIME multiple times, varying sky model parameters to produce multiple model visibilities. χ2 values computed from the model and observed visibilities are used as likelihood values to drive the Bayesian sampling process and select the best sky model. As most of the elements of the RIME and χ2 calculation are independent of one another, they are highly amenable to parallel computation. Additionally, Montblanc caters for iterative RIME evaluation to produce multiple χ2 values. Modified model parameters are transferred to the GPU between each iteration. We implemented Montblanc as a Python package based upon NVIDIA's CUDA architecture. As such, it is easy to extend and implement different pipelines. At present, Montblanc supports point and Gaussian morphologies, but is designed for easy addition of new source profiles. Montblanc's RIME implementation is performant: On an NVIDIA K40, it is approximately 250 times faster than MEQTREES on a dual hexacore Intel E5-2620v2 CPU. Compared to the OSKAR simulator's GPU-implemented RIME components it is 7.7 and 12 times faster on the same K40 for single and double-precision floating point respectively. However, OSKAR's RIME implementation is more general than Montblanc's BIRO-tailored RIME. Theoretical analysis of Montblanc's dominant CUDA kernel suggests that it is memory bound. In practice, profiling shows that is balanced between compute and memory, as much of the data required by the problem is retained in L1 and L2 caches.
Evidence cross-validation and Bayesian inference of MAST plasma equilibria
In this paper, current profiles for plasma discharges on the mega-ampere spherical tokamak are directly calculated from pickup coil, flux loop, and motional-Stark effect observations via methods based in the statistical theory of Bayesian analysis. By representing toroidal plasma current as a series of axisymmetric current beams with rectangular cross-section and inferring the current for each one of these beams, flux-surface geometry and q-profiles are subsequently calculated by elementary application of Biot-Savart's law. The use of this plasma model in the context of Bayesian analysis was pioneered by Svensson and Werner on the joint-European tokamak [Svensson and Werner,Plasma Phys. Controlled Fusion 50(8), 085002 (2008)]. In this framework, linear forward models are used to generate diagnostic predictions, and the probability distribution for the currents in the collection of plasma beams was subsequently calculated directly via application of Bayes' formula. In this work, we introduce a new diagnostic technique to identify and remove outlier observations associated with diagnostics falling out of calibration or suffering from an unidentified malfunction. These modifications enable a good agreement between Bayesian inference of the last-closed flux-surface with other corroborating data, such as that from force balance considerations using EFIT++[Appel et al., ''A unified approach to equilibrium reconstruction'' Proceedings of the 33rd EPS Conference on Plasma Physics (Rome, Italy, 2006)]. In addition, this analysis also yields errors on the plasma current profile and flux-surface geometry as well as directly predicting the Shafranov shift of the plasma core.
Understanding the Scalability of Bayesian Network Inference Using Clique Tree Growth Curves
Mengshoel, Ole J.
2010-01-01
One of the main approaches to performing computation in Bayesian networks (BNs) is clique tree clustering and propagation. The clique tree approach consists of propagation in a clique tree compiled from a Bayesian network, and while it was introduced in the 1980s, there is still a lack of understanding of how clique tree computation time depends on variations in BN size and structure. In this article, we improve this understanding by developing an approach to characterizing clique tree growth as a function of parameters that can be computed in polynomial time from BNs, specifically: (i) the ratio of the number of a BN s non-root nodes to the number of root nodes, and (ii) the expected number of moral edges in their moral graphs. Analytically, we partition the set of cliques in a clique tree into different sets, and introduce a growth curve for the total size of each set. For the special case of bipartite BNs, there are two sets and two growth curves, a mixed clique growth curve and a root clique growth curve. In experiments, where random bipartite BNs generated using the BPART algorithm are studied, we systematically increase the out-degree of the root nodes in bipartite Bayesian networks, by increasing the number of leaf nodes. Surprisingly, root clique growth is well-approximated by Gompertz growth curves, an S-shaped family of curves that has previously been used to describe growth processes in biology, medicine, and neuroscience. We believe that this research improves the understanding of the scaling behavior of clique tree clustering for a certain class of Bayesian networks; presents an aid for trade-off studies of clique tree clustering using growth curves; and ultimately provides a foundation for benchmarking and developing improved BN inference and machine learning algorithms.
Bickel, David R
2011-01-01
In statistical practice, whether a Bayesian or frequentist approach is used in inference depends not only on the availability of prior information but also on the attitude taken toward partial prior information, with frequentists tending to be more cautious than Bayesians. The proposed framework defines that attitude in terms of a specified amount of caution, thereby enabling data analysis at the level of caution desired and on the basis of any prior information. The caution parameter represents the attitude toward partial prior information in much the same way as a loss function represents the attitude toward risk. When there is very little prior information and nonzero caution, the resulting inferences correspond to those of the candidate confidence intervals and p-values that are most similar to the credible intervals and hypothesis probabilities of the specified Bayesian posterior. On the other hand, in the presence of a known physical distribution of the parameter, inferences are based only on the corres...
Dagne, Getachew; Huang, Yangxin
2012-01-01
Censored data are characteristics of many bioassays in HIV/AIDS studies where assays may not be sensitive enough to determine gradations in viral load determination among those below a detectable threshold. Not accounting for such left-censoring appropriately can lead to biased parameter estimates in most data analysis. To properly adjust for left-censoring, this paper presents an extension of the Tobit model for fitting nonlinear dynamic mixed-effects models with skew distributions. Such extensions allow one to specify the conditional distributions for viral load response to account for left-censoring, skewness and heaviness in the tails of the distributions of the response variable. A Bayesian modeling approach via Markov Chain Monte Carlo (MCMC) algorithm is used to estimate model parameters. The proposed methods are illustrated using real data from an HIV/AIDS study. PMID:22992288
Dagne, Getachew; Huang, Yangxin
2016-01-01
Censored data are characteristics of many bioassays in HIV/AIDS studies where assays may not be sensitive enough to determine gradations in viral load determination among those below a detectable threshold. Not accounting for such left-censoring appropriately can lead to biased parameter estimates in most data analysis. To properly adjust for left-censoring, this paper presents an extension of the Tobit model for fitting nonlinear dynamic mixed-effects models with skew distributions. Such extensions allow one to specify the conditional distributions for viral load response to account for left-censoring, skewness and heaviness in the tails of the distributions of the response variable. A Bayesian modeling approach via Markov Chain Monte Carlo (MCMC) algorithm is used to estimate model parameters. The proposed methods are illustrated using real data from an HIV/AIDS study. PMID:22992288
Probabilistic Damage Characterization Using the Computationally-Efficient Bayesian Approach
Warner, James E.; Hochhalter, Jacob D.
2016-01-01
This work presents a computationally-ecient approach for damage determination that quanti es uncertainty in the provided diagnosis. Given strain sensor data that are polluted with measurement errors, Bayesian inference is used to estimate the location, size, and orientation of damage. This approach uses Bayes' Theorem to combine any prior knowledge an analyst may have about the nature of the damage with information provided implicitly by the strain sensor data to form a posterior probability distribution over possible damage states. The unknown damage parameters are then estimated based on samples drawn numerically from this distribution using a Markov Chain Monte Carlo (MCMC) sampling algorithm. Several modi cations are made to the traditional Bayesian inference approach to provide signi cant computational speedup. First, an ecient surrogate model is constructed using sparse grid interpolation to replace a costly nite element model that must otherwise be evaluated for each sample drawn with MCMC. Next, the standard Bayesian posterior distribution is modi ed using a weighted likelihood formulation, which is shown to improve the convergence of the sampling process. Finally, a robust MCMC algorithm, Delayed Rejection Adaptive Metropolis (DRAM), is adopted to sample the probability distribution more eciently. Numerical examples demonstrate that the proposed framework e ectively provides damage estimates with uncertainty quanti cation and can yield orders of magnitude speedup over standard Bayesian approaches.
TYPE Ia SUPERNOVA LIGHT-CURVE INFERENCE: HIERARCHICAL BAYESIAN ANALYSIS IN THE NEAR-INFRARED
We present a comprehensive statistical analysis of the properties of Type Ia supernova (SN Ia) light curves in the near-infrared using recent data from Peters Automated InfraRed Imaging TELescope and the literature. We construct a hierarchical Bayesian framework, incorporating several uncertainties including photometric error, peculiar velocities, dust extinction, and intrinsic variations, for principled and coherent statistical inference. SN Ia light-curve inferences are drawn from the global posterior probability of parameters describing both individual supernovae and the population conditioned on the entire SN Ia NIR data set. The logical structure of the hierarchical model is represented by a directed acyclic graph. Fully Bayesian analysis of the model and data is enabled by an efficient Markov Chain Monte Carlo algorithm exploiting the conditional probabilistic structure using Gibbs sampling. We apply this framework to the JHKs SN Ia light-curve data. A new light-curve model captures the observed J-band light-curve shape variations. The marginal intrinsic variances in peak absolute magnitudes are σ(MJ) = 0.17 ± 0.03, σ(MH) = 0.11 ± 0.03, and σ(MKs) = 0.19 ± 0.04. We describe the first quantitative evidence for correlations between the NIR absolute magnitudes and J-band light-curve shapes, and demonstrate their utility for distance estimation. The average residual in the Hubble diagram for the training set SNe at cz > 2000kms-1 is 0.10 mag. The new application of bootstrap cross-validation to SN Ia light-curve inference tests the sensitivity of the statistical model fit to the finite sample and estimates the prediction error at 0.15 mag. These results demonstrate that SN Ia NIR light curves are as effective as corrected optical light curves, and, because they are less vulnerable to dust absorption, they have great potential as precise and accurate cosmological distance indicators.
Albert, Carlo; Ulzega, Simone; Stoop, Ruedi
2016-04-01
Parameter inference is a fundamental problem in data-driven modeling. Given observed data that is believed to be a realization of some parameterized model, the aim is to find parameter values that are able to explain the observed data. In many situations, the dominant sources of uncertainty must be included into the model for making reliable predictions. This naturally leads to stochastic models. Stochastic models render parameter inference much harder, as the aim then is to find a distribution of likely parameter values. In Bayesian statistics, which is a consistent framework for data-driven learning, this so-called posterior distribution can be used to make probabilistic predictions. We propose a novel, exact, and very efficient approach for generating posterior parameter distributions for stochastic differential equation models calibrated to measured time series. The algorithm is inspired by reinterpreting the posterior distribution as a statistical mechanics partition function of an object akin to a polymer, where the measurements are mapped on heavier beads compared to those of the simulated data. To arrive at distribution samples, we employ a Hamiltonian Monte Carlo approach combined with a multiple time-scale integration. A separation of time scales naturally arises if either the number of measurement points or the number of simulation points becomes large. Furthermore, at least for one-dimensional problems, we can decouple the harmonic modes between measurement points and solve the fastest part of their dynamics analytically. Our approach is applicable to a wide range of inference problems and is highly parallelizable.
Alsing, Justin; Jaffe, Andrew H
2016-01-01
We apply two Bayesian hierarchical inference schemes to infer shear power spectra, shear maps and cosmological parameters from the CFHTLenS weak lensing survey - the first application of this method to data. In the first approach, we sample the joint posterior distribution of the shear maps and power spectra by Gibbs sampling, with minimal model assumptions. In the second approach, we sample the joint posterior of the shear maps and cosmological parameters, providing a new, accurate and principled approach to cosmological parameter inference from cosmic shear data. As a first demonstration on data we perform a 2-bin tomographic analysis to constrain cosmological parameters and investigate the possibility of photometric redshift bias in the CFHTLenS data. Under the baseline $\\Lambda$CDM model we constrain $S_8 = \\sigma_8(\\Omega_\\mathrm{m}/0.3)^{0.5} = 0.67 ^{\\scriptscriptstyle+ 0.03 }_{\\scriptscriptstyle- 0.03 }$ $(68\\%)$, consistent with previous CFHTLenS analysis but in tension with Planck. Adding neutrino m...
von Nessi, G T
2012-01-01
A new method, based on Bayesian analysis, is presented which unifies the inference of plasma equilibria parameters in a Tokamak with the ability to quantify differences between inferred equilibria and Grad-Shafranov force-balance solutions. At the heart of this technique is the new method of observation splitting, which allows multiple forward models to be associated with a single diagnostic observation. This new idea subsequently provides a means by which the the space of GS solutions can be efficiently characterised via a prior distribution. Moreover, by folding force-balance directly into one set of forward models and utilising simple Biot-Savart responses in another, the Bayesian inference of the plasma parameters itself produces an evidence (a normalisation constant of the inferred posterior distribution) which is sensitive to the relative consistency between both sets of models. This evidence can then be used to help determine the relative accuracy of the tested force-balance model across several discha...
A Bayesian method for inferring transmission chains in a partially observed epidemic.
Marzouk, Youssef M.; Ray, Jaideep
2008-10-01
We present a Bayesian approach for estimating transmission chains and rates in the Abakaliki smallpox epidemic of 1967. The epidemic affected 30 individuals in a community of 74; only the dates of appearance of symptoms were recorded. Our model assumes stochastic transmission of the infections over a social network. Distinct binomial random graphs model intra- and inter-compound social connections, while disease transmission over each link is treated as a Poisson process. Link probabilities and rate parameters are objects of inference. Dates of infection and recovery comprise the remaining unknowns. Distributions for smallpox incubation and recovery periods are obtained from historical data. Using Markov chain Monte Carlo, we explore the joint posterior distribution of the scalar parameters and provide an expected connectivity pattern for the social graph and infection pathway.
Improving PWR core simulations by Monte Carlo uncertainty analysis and Bayesian inference
Castro, Emilio; Buss, Oliver; Garcia-Herranz, Nuria; Hoefer, Axel; Porsch, Dieter
2016-01-01
A Monte Carlo-based Bayesian inference model is applied to the prediction of reactor operation parameters of a PWR nuclear power plant. In this non-perturbative framework, high-dimensional covariance information describing the uncertainty of microscopic nuclear data is combined with measured reactor operation data in order to provide statistically sound, well founded uncertainty estimates of integral parameters, such as the boron letdown curve and the burnup-dependent reactor power distribution. The performance of this methodology is assessed in a blind test approach, where we use measurements of a given reactor cycle to improve the prediction of the subsequent cycle. As it turns out, the resulting improvement of the prediction quality is impressive. In particular, the prediction uncertainty of the boron letdown curve, which is of utmost importance for the planning of the reactor cycle length, can be reduced by one order of magnitude by including the boron concentration measurement information of the previous...
Validi, AbdoulAhad
2013-01-01
This study introduces a non-intrusive approach in the context of low-rank separated representation to construct a surrogate of high-dimensional stochastic functions, e.g., PDEs/ODEs, in order to decrease the computational cost of Markov Chain Monte Carlo simulations in Bayesian inference. The surrogate model is constructed via a regularized alternative least-square regression with Tikhonov regularization using a roughening matrix computing the gradient of the solution, in conjunction with a perturbation-based error indicator to detect optimal model complexities. The model approximates a vector of a continuous solution at discrete values of a physical variable. The required number of random realizations to achieve a successful approximation linearly depends on the function dimensionality. The computational cost of the model construction is quadratic in the number of random inputs, which potentially tackles the curse of dimensionality in high-dimensional stochastic functions. Furthermore, this vector valued sep...
Mocapy++ - A toolkit for inference and learning in dynamic Bayesian networks
Hamelryck Thomas
2010-03-01
Full Text Available Abstract Background Mocapy++ is a toolkit for parameter learning and inference in dynamic Bayesian networks (DBNs. It supports a wide range of DBN architectures and probability distributions, including distributions from directional statistics (the statistics of angles, directions and orientations. Results The program package is freely available under the GNU General Public Licence (GPL from SourceForge http://sourceforge.net/projects/mocapy. The package contains the source for building the Mocapy++ library, several usage examples and the user manual. Conclusions Mocapy++ is especially suitable for constructing probabilistic models of biomolecular structure, due to its support for directional statistics. In particular, it supports the Kent distribution on the sphere and the bivariate von Mises distribution on the torus. These distributions have proven useful to formulate probabilistic models of protein and RNA structure in atomic detail.
Hill, T; Minier, V; Burton, M G; Cunningham, M R
2008-01-01
Concatenating data from the millimetre regime to the infrared, we have performed spectral energy distribution modelling for 227 of the 405 millimetre continuum sources of Hill et al. (2005) which are thought to contain young massive stars in the earliest stages of their formation. Three main parameters are extracted from the fits: temperature, mass and luminosity. The method employed was Bayesian inference, which allows a statistically probable range of suitable values for each parameter to be drawn for each individual protostellar candidate. This is the first application of this method to massive star formation. The cumulative distribution plots of the SED modelled parameters in this work indicate that collectively, the sources without methanol maser and/or radio continuum associations (MM-only cores) display similar characteristics to those of high mass star formation regions. Attributing significance to the marginal distinctions between the MM-only cores and the high-mass star formation sample we draw hypo...
Chakraborty, Shubhankar; Roy Chaudhuri, Partha; Das, Prasanta Kr.
2016-07-01
In this communication, a novel optical technique has been proposed for the reconstruction of the shape of a Taylor bubble using measurements from multiple arrays of optical sensors. The deviation of an optical beam passing through the bubble depends on the contour of bubble surface. A theoretical model of the deviation of a beam during the traverse of a Taylor bubble through it has been developed. Using this model and the time history of the deviation captured by the sensor array, the bubble shape has been reconstructed. The reconstruction has been performed using an inverse algorithm based on Bayesian inference technique and Markov chain Monte Carlo sampling algorithm. The reconstructed nose shape has been compared with the true shape, extracted through image processing of high speed images. Finally, an error analysis has been performed to pinpoint the sources of the errors.
Optimal modeling of 1D azimuth correlations in the context of Bayesian inference
De Kock, Michiel B; Trainor, Thomas A
2015-01-01
Analysis and interpretation of spectrum and correlation data from high-energy nuclear collisions is currently controversial because two opposing physics narratives derive contradictory implications from the same data-one narrative claiming collision dynamics is dominated by dijet production and projectile-nucleon fragmentation, the other claiming collision dynamics is dominated by a dense, flowing QCD medium. Opposing interpretations seem to be supported by alternative data models, and current model-comparison schemes are unable to distinguish between them. There is clearly need for a convincing new methodology to break the deadlock. In this study we introduce Bayesian Inference (BI) methods applied to angular correlation data as a basis to evaluate competing data models. For simplicity the data considered are projections of 2D angular correlations onto 1D azimuth from three centrality classes of 200 GeV Au-Au collisions. We consider several data models typical of current model choices, including Fourier seri...
cosmoabc: Likelihood-free inference via Population Monte Carlo Approximate Bayesian Computation
Ishida, E E O; Penna-Lima, M; Cisewski, J; de Souza, R S; Trindade, A M M; Cameron, E
2015-01-01
Approximate Bayesian Computation (ABC) enables parameter inference for complex physical systems in cases where the true likelihood function is unknown, unavailable, or computationally too expensive. It relies on the forward simulation of mock data and comparison between observed and synthetic catalogues. Here we present cosmoabc, a Python ABC sampler featuring a Population Monte Carlo (PMC) variation of the original ABC algorithm, which uses an adaptive importance sampling scheme. The code is very flexible and can be easily coupled to an external simulator, while allowing to incorporate arbitrary distance and prior functions. As an example of practical application, we coupled cosmoabc with the numcosmo library and demonstrate how it can be used to estimate posterior probability distributions over cosmological parameters based on measurements of galaxy clusters number counts without computing the likelihood function. cosmoabc is published under the GPLv3 license on PyPI and GitHub and documentation is availabl...
Giuntini, Michael E.; Giuntini, Ronald E.
1991-01-01
A Bayesian inference process for system logistical planning is presented which provides a method for incorporating actual failures with prediction data for an ongoing and improving reliability estimates. The process uses the Weibull distribution, and provides a means for examining and updating logistical and maintenance support needs.
Bayesian inference in camera trapping studies for a class of spatial capture-recapture models
Royle, J. Andrew; Karanth, K. Ullas; Gopalaswamy, Arjun M.; Kumar, N. Samba
2009-01-01
We develop a class of models for inference about abundance or density using spatial capture-recapture data from studies based on camera trapping and related methods. The model is a hierarchical model composed of two components: a point process model describing the distribution of individuals in space (or their home range centers) and a model describing the observation of individuals in traps. We suppose that trap- and individual-specific capture probabilities are a function of distance between individual home range centers and trap locations. We show that the models can be regarded as generalized linear mixed models, where the individual home range centers are random effects. We adopt a Bayesian framework for inference under these models using a formulation based on data augmentation. We apply the models to camera trapping data on tigers from the Nagarahole Reserve, India, collected over 48 nights in 2006. For this study, 120 camera locations were used, but cameras were only operational at 30 locations during any given sample occasion. Movement of traps is common in many camera-trapping studies and represents an important feature of the observation model that we address explicitly in our application.
Bayesian inference of biochemical kinetic parameters using the linear noise approximation
Finkenstädt Bärbel
2009-10-01
Full Text Available Abstract Background Fluorescent and luminescent gene reporters allow us to dynamically quantify changes in molecular species concentration over time on the single cell level. The mathematical modeling of their interaction through multivariate dynamical models requires the deveopment of effective statistical methods to calibrate such models against available data. Given the prevalence of stochasticity and noise in biochemical systems inference for stochastic models is of special interest. In this paper we present a simple and computationally efficient algorithm for the estimation of biochemical kinetic parameters from gene reporter data. Results We use the linear noise approximation to model biochemical reactions through a stochastic dynamic model which essentially approximates a diffusion model by an ordinary differential equation model with an appropriately defined noise process. An explicit formula for the likelihood function can be derived allowing for computationally efficient parameter estimation. The proposed algorithm is embedded in a Bayesian framework and inference is performed using Markov chain Monte Carlo. Conclusion The major advantage of the method is that in contrast to the more established diffusion approximation based methods the computationally costly methods of data augmentation are not necessary. Our approach also allows for unobserved variables and measurement error. The application of the method to both simulated and experimental data shows that the proposed methodology provides a useful alternative to diffusion approximation based methods.
ESTIMATE OF THE HYPSOMETRIC RELATIONSHIP WITH NONLINEAR MODELS FITTED BY EMPIRICAL BAYESIAN METHODS
Monica Fabiana Bento Moreira
2015-09-01
Full Text Available In this paper we propose a Bayesian approach to solve the inference problem with restriction on parameters, regarding to nonlinear models used to represent the hypsometric relationship in clones of Eucalyptus sp. The Bayesian estimates are calculated using Monte Carlo Markov Chain (MCMC method. The proposed method was applied to different groups of actual data from which two were selected to show the results. These results were compared to the results achieved by the minimum square method, highlighting the superiority of the Bayesian approach, since this approach always generate the biologically consistent results for hipsometric relationship.
Storz, Jay F; Beaumont, Mark A; Alberts, Susan C
2002-11-01
The purpose of this study was to test for evidence that savannah baboons (Papio cynocephalus) underwent a population expansion in concert with a hypothesized expansion of African human and chimpanzee populations during the late Pleistocene. The rationale is that any type of environmental event sufficient to cause simultaneous population expansions in African humans and chimpanzees would also be expected to affect other codistributed mammals. To test for genetic evidence of population expansion or contraction, we performed a coalescent analysis of multilocus microsatellite data using a hierarchical Bayesian model. Markov chain Monte Carlo (MCMC) simulations were used to estimate the posterior probability density of demographic and genealogical parameters. The model was designed to allow interlocus variation in mutational and demographic parameters, which made it possible to detect aberrant patterns of variation at individual loci that could result from heterogeneity in mutational dynamics or from the effects of selection at linked sites. Results of the MCMC simulations were consistent with zero variance in demographic parameters among loci, but there was evidence for a 10- to 20-fold difference in mutation rate between the most slowly and most rapidly evolving loci. Results of the model provided strong evidence that savannah baboons have undergone a long-term historical decline in population size. The mode of the highest posterior density for the joint distribution of current and ancestral population size indicated a roughly eightfold contraction over the past 1,000 to 250,000 years. These results indicate that savannah baboons apparently did not share a common demographic history with other codistributed primate species. PMID:12411607
A Hamiltonian Monte–Carlo method for Bayesian inference of supermassive black hole binaries
We investigate the use of a Hamiltonian Monte–Carlo to map out the posterior density function for supermassive black hole binaries. While previous Markov Chain Monte–Carlo (MCMC) methods, such as Metropolis–Hastings MCMC, have been successfully employed for a number of different gravitational wave sources, these methods are essentially random walk algorithms. The Hamiltonian Monte–Carlo treats the inverse likelihood surface as a ‘gravitational potential’ and by introducing canonical positions and momenta, dynamically evolves the Markov chain by solving Hamilton's equations of motion. This method is not as widely used as other MCMC algorithms due to the necessity of calculating gradients of the log-likelihood, which for most applications results in a bottleneck that makes the algorithm computationally prohibitive. We circumvent this problem by using accepted initial phase-space trajectory points to analytically fit for each of the individual gradients. Eliminating the waveform generation needed for the numerical derivatives reduces the total number of required templates for a 106 iteration chain from ∼109 to ∼106. The result is in an implementation of the Hamiltonian Monte–Carlo that is faster, and more efficient by a factor of approximately the dimension of the parameter space, than a Hessian MCMC. (paper)
Reversible Jump MCMC Simulated Annealing for Neural Networks
Andrieu, Christophe; De Freitas, Nando; Doucet, Arnaud
2013-01-01
We propose a novel reversible jump Markov chain Monte Carlo (MCMC) simulated annealing algorithm to optimize radial basis function (RBF) networks. This algorithm enables us to maximize the joint posterior distribution of the network parameters and the number of basis functions. It performs a global search in the joint space of the parameters and number of parameters, thereby surmounting the problem of local minima. We also show that by calibrating a Bayesian model, we can obtain the classical...
Bazin, Eric; Dawson, Kevin J; Beaumont, Mark A
2010-06-01
We address the problem of finding evidence of natural selection from genetic data, accounting for the confounding effects of demographic history. In the absence of natural selection, gene genealogies should all be sampled from the same underlying distribution, often approximated by a coalescent model. Selection at a particular locus will lead to a modified genealogy, and this motivates a number of recent approaches for detecting the effects of natural selection in the genome as "outliers" under some models. The demographic history of a population affects the sampling distribution of genealogies, and therefore the observed genotypes and the classification of outliers. Since we cannot see genealogies directly, we have to infer them from the observed data under some model of mutation and demography. Thus the accuracy of an outlier-based approach depends to a greater or a lesser extent on the uncertainty about the demographic and mutational model. A natural modeling framework for this type of problem is provided by Bayesian hierarchical models, in which parameters, such as mutation rates and selection coefficients, are allowed to vary across loci. It has proved quite difficult computationally to implement fully probabilistic genealogical models with complex demographies, and this has motivated the development of approximations such as approximate Bayesian computation (ABC). In ABC the data are compressed into summary statistics, and computation of the likelihood function is replaced by simulation of data under the model. In a hierarchical setting one may be interested both in hyperparameters and parameters, and there may be very many of the latter--for example, in a genetic model, these may be parameters describing each of many loci or populations. This poses a problem for ABC in that one then requires summary statistics for each locus, which, if used naively, leads to a consequent difficulty in conditional density estimation. We develop a general method for applying
Raue, Andreas; Theis, Fabian Joachim; Timmer, Jens
2012-01-01
Increasingly complex applications involve large datasets in combination with non-linear and high dimensional mathematical models. In this context, statistical inference is a challenging issue that calls for pragmatic approaches that take advantage of both Bayesian and frequentist methods. The elegance of Bayesian methodology is founded in the propagation of information content provided by experimental data and prior assumptions to the posterior probability distribution of model predictions. However, for complex applications experimental data and prior assumptions potentially constrain the posterior probability distribution insufficiently. In these situations Bayesian Markov chain Monte Carlo sampling can be infeasible. From a frequentist point of view insufficient experimental data and prior assumptions can be interpreted as non-identifiability. The profile likelihood approach offers to detect and to resolve non-identifiability by experimental design iteratively. Therefore, it allows one to better constrain t...
Cuevas Rivera, Dario; Bitzer, Sebastian; Kiebel, Stefan J
2015-10-01
The olfactory information that is received by the insect brain is encoded in the form of spatiotemporal patterns in the projection neurons of the antennal lobe. These dense and overlapping patterns are transformed into a sparse code in Kenyon cells in the mushroom body. Although it is clear that this sparse code is the basis for rapid categorization of odors, it is yet unclear how the sparse code in Kenyon cells is computed and what information it represents. Here we show that this computation can be modeled by sequential firing rate patterns using Lotka-Volterra equations and Bayesian online inference. This new model can be understood as an 'intelligent coincidence detector', which robustly and dynamically encodes the presence of specific odor features. We found that the model is able to qualitatively reproduce experimentally observed activity in both the projection neurons and the Kenyon cells. In particular, the model explains mechanistically how sparse activity in the Kenyon cells arises from the dense code in the projection neurons. The odor classification performance of the model proved to be robust against noise and time jitter in the observed input sequences. As in recent experimental results, we found that recognition of an odor happened very early during stimulus presentation in the model. Critically, by using the model, we found surprising but simple computational explanations for several experimental phenomena. PMID:26451888
Dario Cuevas Rivera
2015-10-01
Full Text Available The olfactory information that is received by the insect brain is encoded in the form of spatiotemporal patterns in the projection neurons of the antennal lobe. These dense and overlapping patterns are transformed into a sparse code in Kenyon cells in the mushroom body. Although it is clear that this sparse code is the basis for rapid categorization of odors, it is yet unclear how the sparse code in Kenyon cells is computed and what information it represents. Here we show that this computation can be modeled by sequential firing rate patterns using Lotka-Volterra equations and Bayesian online inference. This new model can be understood as an 'intelligent coincidence detector', which robustly and dynamically encodes the presence of specific odor features. We found that the model is able to qualitatively reproduce experimentally observed activity in both the projection neurons and the Kenyon cells. In particular, the model explains mechanistically how sparse activity in the Kenyon cells arises from the dense code in the projection neurons. The odor classification performance of the model proved to be robust against noise and time jitter in the observed input sequences. As in recent experimental results, we found that recognition of an odor happened very early during stimulus presentation in the model. Critically, by using the model, we found surprising but simple computational explanations for several experimental phenomena.
Bayesian model comparison and parameter inference in systems biology using nested sampling.
Pullen, Nick; Morris, Richard J
2014-01-01
Inferring parameters for models of biological processes is a current challenge in systems biology, as is the related problem of comparing competing models that explain the data. In this work we apply Skilling's nested sampling to address both of these problems. Nested sampling is a Bayesian method for exploring parameter space that transforms a multi-dimensional integral to a 1D integration over likelihood space. This approach focuses on the computation of the marginal likelihood or evidence. The ratio of evidences of different models leads to the Bayes factor, which can be used for model comparison. We demonstrate how nested sampling can be used to reverse-engineer a system's behaviour whilst accounting for the uncertainty in the results. The effect of missing initial conditions of the variables as well as unknown parameters is investigated. We show how the evidence and the model ranking can change as a function of the available data. Furthermore, the addition of data from extra variables of the system can deliver more information for model comparison than increasing the data from one variable, thus providing a basis for experimental design. PMID:24523891
Bayesian model comparison and parameter inference in systems biology using nested sampling.
Nick Pullen
Full Text Available Inferring parameters for models of biological processes is a current challenge in systems biology, as is the related problem of comparing competing models that explain the data. In this work we apply Skilling's nested sampling to address both of these problems. Nested sampling is a Bayesian method for exploring parameter space that transforms a multi-dimensional integral to a 1D integration over likelihood space. This approach focuses on the computation of the marginal likelihood or evidence. The ratio of evidences of different models leads to the Bayes factor, which can be used for model comparison. We demonstrate how nested sampling can be used to reverse-engineer a system's behaviour whilst accounting for the uncertainty in the results. The effect of missing initial conditions of the variables as well as unknown parameters is investigated. We show how the evidence and the model ranking can change as a function of the available data. Furthermore, the addition of data from extra variables of the system can deliver more information for model comparison than increasing the data from one variable, thus providing a basis for experimental design.
Bayesian Inference of the Evolution of a Phenotype Distribution on a Phylogenetic Tree
Ansari, M. Azim; Didelot, Xavier
2016-01-01
The distribution of a phenotype on a phylogenetic tree is often a quantity of interest. Many phenotypes have imperfect heritability, so that a measurement of the phenotype for an individual can be thought of as a single realization from the phenotype distribution of that individual. If all individuals in a phylogeny had the same phenotype distribution, measured phenotypes would be randomly distributed on the tree leaves. This is, however, often not the case, implying that the phenotype distribution evolves over time. Here we propose a new model based on this principle of evolving phenotype distribution on the branches of a phylogeny, which is different from ancestral state reconstruction where the phenotype itself is assumed to evolve. We develop an efficient Bayesian inference method to estimate the parameters of our model and to test the evidence for changes in the phenotype distribution. We use multiple simulated data sets to show that our algorithm has good sensitivity and specificity properties. Since our method identifies branches on the tree on which the phenotype distribution has changed, it is able to break down a tree into components for which this distribution is unique and constant. We present two applications of our method, one investigating the association between HIV genetic variation and human leukocyte antigen and the other studying host range distribution in a lineage of Salmonella enterica, and we discuss many other potential applications. PMID:27412711
This study introduces a non-intrusive approach in the context of low-rank separated representation to construct a surrogate of high-dimensional stochastic functions, e.g., PDEs/ODEs, in order to decrease the computational cost of Markov Chain Monte Carlo simulations in Bayesian inference. The surrogate model is constructed via a regularized alternative least-square regression with Tikhonov regularization using a roughening matrix computing the gradient of the solution, in conjunction with a perturbation-based error indicator to detect optimal model complexities. The model approximates a vector of a continuous solution at discrete values of a physical variable. The required number of random realizations to achieve a successful approximation linearly depends on the function dimensionality. The computational cost of the model construction is quadratic in the number of random inputs, which potentially tackles the curse of dimensionality in high-dimensional stochastic functions. Furthermore, this vector-valued separated representation-based model, in comparison to the available scalar-valued case, leads to a significant reduction in the cost of approximation by an order of magnitude equal to the vector size. The performance of the method is studied through its application to three numerical examples including a 41-dimensional elliptic PDE and a 21-dimensional cavity flow
We have developed a Bayesian approach to the analysis of neural electromagnetic (MEG/EEG) data that can incorporate or fuse information from other imaging modalities and addresses the ill-posed inverse problem by sarnpliig the many different solutions which could have produced the given data. From these samples one can draw probabilistic inferences about regions of activation. Our source model assumes a variable number of variable size cortical regions of stimulus-correlated activity. An active region consists of locations on the cortical surf ace, within a sphere centered on some location in cortex. The number and radi of active regions can vary to defined maximum values. The goal of the analysis is to determine the posterior probability distribution for the set of parameters that govern the number, location, and extent of active regions. Markov Chain Monte Carlo is used to generate a large sample of sets of parameters distributed according to the posterior distribution. This sample is representative of the many different source distributions that could account for given data, and allows identification of probable (i.e. consistent) features across solutions. Examples of the use of this analysis technique with both simulated and empirical MEG data are presented
George, J.S.; Schmidt, D.M.; Wood, C.C.
1999-02-01
We have developed a Bayesian approach to the analysis of neural electromagnetic (MEG/EEG) data that can incorporate or fuse information from other imaging modalities and addresses the ill-posed inverse problem by sarnpliig the many different solutions which could have produced the given data. From these samples one can draw probabilistic inferences about regions of activation. Our source model assumes a variable number of variable size cortical regions of stimulus-correlated activity. An active region consists of locations on the cortical surf ace, within a sphere centered on some location in cortex. The number and radi of active regions can vary to defined maximum values. The goal of the analysis is to determine the posterior probability distribution for the set of parameters that govern the number, location, and extent of active regions. Markov Chain Monte Carlo is used to generate a large sample of sets of parameters distributed according to the posterior distribution. This sample is representative of the many different source distributions that could account for given data, and allows identification of probable (i.e. consistent) features across solutions. Examples of the use of this analysis technique with both simulated and empirical MEG data are presented.
Wang, Dong; Tsui, Kwok-Leung; Zhou, Qiang
2016-05-01
Rolling element bearings are commonly used in machines to provide support for rotating shafts. Bearing failures may cause unexpected machine breakdowns and increase economic cost. To prevent machine breakdowns and reduce unnecessary economic loss, bearing faults should be detected as early as possible. Because wavelet transform can be used to highlight impulses caused by localized bearing faults, wavelet transform has been widely investigated and proven to be one of the most effective and efficient methods for bearing fault diagnosis. In this paper, a new Gauss-Hermite integration based Bayesian inference method is proposed to estimate the posterior distribution of wavelet parameters. The innovations of this paper are illustrated as follows. Firstly, a non-linear state space model of wavelet parameters is constructed to describe the relationship between wavelet parameters and hypothetical measurements. Secondly, the joint posterior probability density function of wavelet parameters and hypothetical measurements is assumed to follow a joint Gaussian distribution so as to generate Gaussian perturbations for the state space model. Thirdly, Gauss-Hermite integration is introduced to analytically predict and update moments of the joint Gaussian distribution, from which optimal wavelet parameters are derived. At last, an optimal wavelet filtering is conducted to extract bearing fault features and thus identify localized bearing faults. Two instances are investigated to illustrate how the proposed method works. Two comparisons with the fast kurtogram are used to demonstrate that the proposed method can achieve better visual inspection performances than the fast kurtogram.
Duggento, Andrea; McClintock, Peter V E; Stefanovska, Aneta
2012-01-01
Living systems have time-evolving interactions that, until recently, could not be identified accurately from recorded time series in the presence of noise. Stankovski et al. (Phys. Rev. Lett. 109 024101, 2012) introduced a method based on dynamical Bayesian inference that facilitates the simultaneous detection of time-varying synchronization, directionality of influence, and coupling functions. It can distinguish unsynchronized dynamics from noise-induced phase slips. The method is based on phase dynamics, with Bayesian inference of the time- evolving parameters being achieved by shaping the prior densities to incorporate knowledge of previous samples. We now present the method in detail using numerically-generated data, data from an analog electronic circuit, and cardio-respiratory data. We also generalize the method to encompass networks of interacting oscillators and thus demonstrate its applicability to small-scale networks.
Sudano, Mateus José; Crespilho, André Maciel; Fernandes, Claudia Barbosa; Junior, Alicio Martins; Papa, Frederico Ozanam; Rodrigues, Josemar; Machado, Rui; Landim-Alvarenga, Fernanda Da Cruz
2011-01-01
The objective of this experiment was to test in vitro embryo production (IVP) as a tool to estimate fertility performance in zebu bulls using Bayesian inference statistics. Oocytes were matured and fertilized in vitro using sperm cells from three different Zebu bulls (V, T, and G). The three bulls presented similar results with regard to pronuclear formation and blastocyst formation rates. However, the cleavage rates were different between bulls. The estimated conception rates based on combined data of cleavage and blastocyst formation were very similar to the true conception rates observed for the same bulls after a fixed-time artificial insemination program. Moreover, even when we used cleavage rate data only or blastocyst formation data only, the estimated conception rates were still close to the true conception rates. We conclude that Bayesian inference is an effective statistical procedure to estimate in vivo bull fertility using data from IVP. PMID:21547211
Hug, Sabine Carolin
2015-01-01
In this thesis we use differential equations for mathematically representing biological processes. For this we have to infer the associated parameters for fitting the differential equations to measurement data. If the structure of the ODE itself is uncertain, model selection methods have to be applied. We refine several existing Bayesian methods, ranging from an adaptive scheme for the computation of high-dimensional integrals to multi-chain Metropolis-Hastings algorithms for high-dimensional...
A surrogate model enables a Bayesian approach to the inverse problem of scatterometry
Scatterometry is an indirect optical method for the determination of photomask geometry parameters from scattered light intensities by solving an inverse problem. The Bayesian approach is a powerful method to solve the inverse problem. In the Bayesian framework estimates of parameters and associated uncertainties are obtained from posterior distributions. The determination the probability distribution is typically based on Markov chain Monte Carlo (MCMC) methods. However, in scatterometry the evaluation of MCMC steps require solutions of partial differential equations that are computationally expensive and application of MCMC methods is thus impractical. In this article we introduce a surrogate model for scatterometry based on polynomial chaos that can be treated by Bayesian inference. We compare the results of the surrogate model with rigorous finite element simulations and demonstrate its convergence. The accuracy reaches a value of lower than one percent for a sufficient fine mesh and the speed up amounts more than two order of magnitudes. Furthermore, we apply the surrogate model to MCMC calculations and we reconstruct geometry parameters of a photomask
MCMC for Wind Power Simulation
Papaefthymiou, G.; Klöckl, B.
2008-01-01
This paper contributes a Markov chain Monte Carlo (MCMC) method for the direct generation of synthetic time series of wind power output. It is shown that obtaining a stochastic model directly in the wind power domain leads to reduced number of states and to lower order of the Markov chain at equal p
Bayesian inference along Markov Chain Monte Carlo approach for PWR core loading pattern optimization
Highlights: ► The BIMCMC method performs very well and is comparable to GA and PSO techniques. ► The potential of the technique is very well for optimization. ► It is observed that the performance of the method is quite adequate. ► The BIMCMC is very easy to implement. -- Abstract: Despite remarkable progress in optimization procedures, inherent complexities in nuclear reactor structure and strong interdependence among the fundamental indices namely, economic, neutronic, thermo-hydraulic and environmental effects make it necessary to evaluate the most efficient arrangement of a reactor core. In this paper a reactor core reloading technique based on Bayesian inference along Markov Chain Monte Carlo, BIMCMC, is addressed in the context of obtaining an optimal configuration of fuel assemblies in reactor cores. The Markov Chain Monte Carlo with Metropolis–Hastings algorithm has been applied for sampling variable and its acceptance. The proposed algorithm can be used for in-core fuel management optimization problems in pressurized water reactors. Considerable work has been expended for loading pattern optimization, but no preferred approach has yet emerged. To evaluate the proposed technique, increasing the effective multiplication factor Keff of a WWER-1000 core along flattening power with keeping power peaking factor below a specific limit as a first test case and flattening of power as a second test case are considered as objective functions; although other variables such as burn up and cycle length can also be taken into account. The results, convergence rate and reliability of the new method are compared to published data resulting from particle swarm optimization and genetic algorithm; the outcome is quite promising and demonstrating the potential of the technique very well for optimization applications in the nuclear engineering field.
Kang, Seongkeun; Seong, Poong Hyun [Korea Advanced Institute of Science and Technology, Daejeon (Korea, Republic of)
2014-05-15
The purpose of this paper is to confirm if Bayesian inference can properly reflect the situation awareness of real human operators, and find the difference between the situation of ideal and practical operators, and investigate the factors which contributes to those difference. As a results, human can not think like computer. If human can memorize all the information, and their thinking process is same to the CPU of computer, the results of these two experiments come out more than 99%. However the probability of finding right malfunction by humans are only 64.52% in simple experiment, and 51.61% in complex experiment. Cognition is the mental processing that includes the attention of working memory, comprehending and producing language, calculating, reasoning, problem solving, and decision making. There are many reasons why human thinking process is different with computer, but in this experiment, we suggest that the working memory is the most important factor. Humans have limited working memory which has only seven chunks capacity. These seven chunks are called magic number. If there are more than seven sequential information, people start to forget the previous information because their working memory capacity is running over. We can check how much working memory affects to the result through the simple experiment. Then what if we neglect the effect of working memory? The total number of subjects who have incorrect memory is 7 (subject 3, 5, 6, 7, 8, 15, 25). They could find the right malfunction if the memory hadn't changed because of lack of working memory. Then the probability of find correct malfunction will be increased to 87.10% from 64.52%. Complex experiment has similar result. In this case, eight subjects(1, 5, 8, 9, 15, 17, 18, 30) had changed the memory, and it affects to find the right malfunction. Considering it, then the probability would be (16+8)/31 = 77.42%.
The purpose of this paper is to confirm if Bayesian inference can properly reflect the situation awareness of real human operators, and find the difference between the situation of ideal and practical operators, and investigate the factors which contributes to those difference. As a results, human can not think like computer. If human can memorize all the information, and their thinking process is same to the CPU of computer, the results of these two experiments come out more than 99%. However the probability of finding right malfunction by humans are only 64.52% in simple experiment, and 51.61% in complex experiment. Cognition is the mental processing that includes the attention of working memory, comprehending and producing language, calculating, reasoning, problem solving, and decision making. There are many reasons why human thinking process is different with computer, but in this experiment, we suggest that the working memory is the most important factor. Humans have limited working memory which has only seven chunks capacity. These seven chunks are called magic number. If there are more than seven sequential information, people start to forget the previous information because their working memory capacity is running over. We can check how much working memory affects to the result through the simple experiment. Then what if we neglect the effect of working memory? The total number of subjects who have incorrect memory is 7 (subject 3, 5, 6, 7, 8, 15, 25). They could find the right malfunction if the memory hadn't changed because of lack of working memory. Then the probability of find correct malfunction will be increased to 87.10% from 64.52%. Complex experiment has similar result. In this case, eight subjects(1, 5, 8, 9, 15, 17, 18, 30) had changed the memory, and it affects to find the right malfunction. Considering it, then the probability would be (16+8)/31 = 77.42%
Multi-scale inference of interaction rules in animal groups using Bayesian model selection.
Richard P Mann
Full Text Available Inference of interaction rules of animals moving in groups usually relies on an analysis of large scale system behaviour. Models are tuned through repeated simulation until they match the observed behaviour. More recent work has used the fine scale motions of animals to validate and fit the rules of interaction of animals in groups. Here, we use a Bayesian methodology to compare a variety of models to the collective motion of glass prawns (Paratya australiensis. We show that these exhibit a stereotypical 'phase transition', whereby an increase in density leads to the onset of collective motion in one direction. We fit models to this data, which range from: a mean-field model where all prawns interact globally; to a spatial Markovian model where prawns are self-propelled particles influenced only by the current positions and directions of their neighbours; up to non-Markovian models where prawns have 'memory' of previous interactions, integrating their experiences over time when deciding to change behaviour. We show that the mean-field model fits the large scale behaviour of the system, but does not capture the observed locality of interactions. Traditional self-propelled particle models fail to capture the fine scale dynamics of the system. The most sophisticated model, the non-Markovian model, provides a good match to the data at both the fine scale and in terms of reproducing global dynamics, while maintaining a biologically plausible perceptual range. We conclude that prawns' movements are influenced by not just the current direction of nearby conspecifics, but also those encountered in the recent past. Given the simplicity of prawns as a study system our research suggests that self-propelled particle models of collective motion should, if they are to be realistic at multiple biological scales, include memory of previous interactions and other non-Markovian effects.
Multi-scale inference of interaction rules in animal groups using Bayesian model selection.
Richard P Mann
2012-01-01
Full Text Available Inference of interaction rules of animals moving in groups usually relies on an analysis of large scale system behaviour. Models are tuned through repeated simulation until they match the observed behaviour. More recent work has used the fine scale motions of animals to validate and fit the rules of interaction of animals in groups. Here, we use a Bayesian methodology to compare a variety of models to the collective motion of glass prawns (Paratya australiensis. We show that these exhibit a stereotypical 'phase transition', whereby an increase in density leads to the onset of collective motion in one direction. We fit models to this data, which range from: a mean-field model where all prawns interact globally; to a spatial Markovian model where prawns are self-propelled particles influenced only by the current positions and directions of their neighbours; up to non-Markovian models where prawns have 'memory' of previous interactions, integrating their experiences over time when deciding to change behaviour. We show that the mean-field model fits the large scale behaviour of the system, but does not capture fine scale rules of interaction, which are primarily mediated by physical contact. Conversely, the Markovian self-propelled particle model captures the fine scale rules of interaction but fails to reproduce global dynamics. The most sophisticated model, the non-Markovian model, provides a good match to the data at both the fine scale and in terms of reproducing global dynamics. We conclude that prawns' movements are influenced by not just the current direction of nearby conspecifics, but also those encountered in the recent past. Given the simplicity of prawns as a study system our research suggests that self-propelled particle models of collective motion should, if they are to be realistic at multiple biological scales, include memory of previous interactions and other non-Markovian effects.
HASSET: a probability event tree tool to evaluate future volcanic scenarios using Bayesian inference
Sobradelo, Rosa; Bartolini, Stefania; Martí, Joan
2014-01-01
Event tree structures constitute one of the most useful and necessary tools in modern volcanology for assessment of hazards from future volcanic scenarios (those that culminate in an eruptive event as well as those that do not). They are particularly relevant for evaluation of long- and short-term probabilities of occurrence of possible volcanic scenarios and their potential impacts on urbanized areas. In this paper, we introduce Hazard Assessment Event Tree (HASSET), a probability tool, built on an event tree structure that uses Bayesian inference to estimate the probability of occurrence of a future volcanic scenario and to evaluate the most relevant sources of uncertainty from the corresponding volcanic system. HASSET includes hazard assessment of noneruptive and nonmagmatic volcanic scenarios, that is, episodes of unrest that do not evolve into volcanic eruption but have an associated volcanic hazard (e.g., sector collapse and phreatic explosion), as well as unrest episodes triggered by external triggers rather than the magmatic system alone. Additionally, HASSET introduces the Delta method to assess precision of the probability estimates, by reporting a 1 standard deviation variability interval around the expected value for each scenario. HASSET is presented as a free software package in the form of a plug-in for the open source geographic information system Quantum Gis (QGIS), providing a graphically supported computation of the event tree structure in an interactive and user-friendly way. We also include further in-depth explanations for each node together with an application of HASSET to Teide-Pico Viejo volcanic complex (Spain).
MCMC joint separation and segmentation of hidden Markov fields
Snoussi, H; Snoussi, Hichem; Mohammad-Djafari, Ali
2002-01-01
In this contribution, we consider the problem of the blind separation of noisy instantaneously mixed images. The images are modelized by hidden Markov fields with unknown parameters. Given the observed images, we give a Bayesian formulation and we propose to solve the resulting data augmentation problem by implementing a Monte Carlo Markov Chain (MCMC) procedure. We separate the unknown variables into two categories: 1. The parameters of interest which are the mixing matrix, the noise covariance and the parameters of the sources distributions. 2. The hidden variables which are the unobserved sources and the unobserved pixels classification labels. The proposed algorithm provides in the stationary regime samples drawn from the posterior distributions of all the variables involved in the problem leading to a flexibility in the cost function choice. We discuss and characterize some problems of non identifiability and degeneracies of the parameters likelihood and the behavior of the MCMC algorithm in this case. F...
Fang-Rong Yan
Full Text Available This article provides a fully bayesian approach for modeling of single-dose and complete pharmacokinetic data in a population pharmacokinetic (PK model. To overcome the impact of outliers and the difficulty of computation, a generalized linear model is chosen with the hypothesis that the errors follow a multivariate Student t distribution which is a heavy-tailed distribution. The aim of this study is to investigate and implement the performance of the multivariate t distribution to analyze population pharmacokinetic data. Bayesian predictive inferences and the Metropolis-Hastings algorithm schemes are used to process the intractable posterior integration. The precision and accuracy of the proposed model are illustrated by the simulating data and a real example of theophylline data.
Analysis of trend changes in Northern African palaeo-climate by using Bayesian inference
Schütz, Nadine; Trauth, Martin H.; Holschneider, Matthias
2010-05-01
Climate variability of Northern Africa is of high interest due to climate-evolutionary linkages under study. The reconstruction of the palaeo-climate over long time scales, including the expected linkages (> 3 Ma), is mainly accessible by proxy data from deep sea drilling cores. By concentrating on published data sets, we try to decipher rhythms and trends to detect correlations between different proxy time series by advanced mathematical methods. Our preliminary data is dust concentration, as an indicator for climatic changes such as humidity, from the ODP sites 659, 721 and 967 situated around Northern Africa. Our interest is in challenging the available time series with advanced statistical methods to detect significant trend changes and to compare different model assumptions. For that purpose, we want to avoid the rescaling of the time axis to obtain equidistant time steps for filtering methods. Additionally we demand an plausible description of the errors for the estimated parameters, in terms of confidence intervals. Finally, depending on what model we restrict on, we also want an insight in the parameter structure of the assumed models. To gain this information, we focus on Bayesian inference by formulating the problem as a linear mixed model, so that the expectation and deviation are of linear structure. By using the Bayesian method we can formulate the posteriori density as a function of the model parameters and calculate this probability density in the parameter space. Depending which parameters are of interest, we analytically and numerically marginalize the posteriori with respect to the remaining parameters of less interest. We apply a simple linear mixed model to calculate the posteriori densities of the ODP sites 659 and 721 concerning the last 5 Ma at maximum. From preliminary calculations on these data sets, we can confirm results gained by the method of breakfit regression combined with block bootstrapping ([1]). We obtain a significant change
Edson Sandoval-Castellanos
Full Text Available Inference of population demographic history has vastly improved in recent years due to a number of technological and theoretical advances including the use of ancient DNA. Approximate Bayesian computation (ABC stands among the most promising methods due to its simple theoretical fundament and exceptional flexibility. However, limited availability of user-friendly programs that perform ABC analysis renders it difficult to implement, and hence programming skills are frequently required. In addition, there is limited availability of programs able to deal with heterochronous data. Here we present the software BaySICS: Bayesian Statistical Inference of Coalescent Simulations. BaySICS provides an integrated and user-friendly platform that performs ABC analyses by means of coalescent simulations from DNA sequence data. It estimates historical demographic population parameters and performs hypothesis testing by means of Bayes factors obtained from model comparisons. Although providing specific features that improve inference from datasets with heterochronous data, BaySICS also has several capabilities making it a suitable tool for analysing contemporary genetic datasets. Those capabilities include joint analysis of independent tables, a graphical interface and the implementation of Markov-chain Monte Carlo without likelihoods.
Sankararaman, Shankar
2016-01-01
This paper presents a computational framework for uncertainty characterization and propagation, and sensitivity analysis under the presence of aleatory and epistemic un- certainty, and develops a rigorous methodology for efficient refinement of epistemic un- certainty by identifying important epistemic variables that significantly affect the overall performance of an engineering system. The proposed methodology is illustrated using the NASA Langley Uncertainty Quantification Challenge (NASA-LUQC) problem that deals with uncertainty analysis of a generic transport model (GTM). First, Bayesian inference is used to infer subsystem-level epistemic quantities using the subsystem-level model and corresponding data. Second, tools of variance-based global sensitivity analysis are used to identify four important epistemic variables (this limitation specified in the NASA-LUQC is reflective of practical engineering situations where not all epistemic variables can be refined due to time/budget constraints) that significantly affect system-level performance. The most significant contribution of this paper is the development of the sequential refine- ment methodology, where epistemic variables for refinement are not identified all-at-once. Instead, only one variable is first identified, and then, Bayesian inference and global sensi- tivity calculations are repeated to identify the next important variable. This procedure is continued until all 4 variables are identified and the refinement in the system-level perfor- mance is computed. The advantages of the proposed sequential refinement methodology over the all-at-once uncertainty refinement approach are explained, and then applied to the NASA Langley Uncertainty Quantification Challenge problem.