Search results for "markov chain"

Markov chain Monte Carlo importance samplers for Bayesian models with intractable likelihoods

2019

Markov chain Monte Carlo (MCMC) is an approach to parameter inference in Bayesian models that is based on computing ergodic averages formed from a Markov chain targeting the Bayesian posterior probability. We consider the efficient use of an approximation within the Markov chain, with subsequent importance sampling (IS) correction of the Markov chain inexact output, leading to asymptotically exact inference. We detail convergence and central limit theorems for the resulting MCMC-IS estimators. We also consider the case where the approximate Markov chain is pseudo-marginal, requiring unbiased estimators for its approximate marginal target. Convergence results with asymptotic variance formula…

ℓ1-Penalized Methods in High-Dimensional Gaussian Markov Random Fields

2016

In the last 20 years, we have witnessed the dramatic development of new data acquisition technologies allowing to collect massive amount of data with relatively low cost. is new feature leads Donoho to define the twenty-first century as the century of data. A major characteristic of this modern data set is that the number of measured variables is larger than the sample size; the word high-dimensional data analysis is referred to the statistical methods developed to make inference with this new kind of data. This chapter is devoted to the study of some of the most recent ℓ1-penalized methods proposed in the literature to make sparse inference in a Gaussian Markov random field (GMRF) defined …

Matemātika

2005

Communities of Local Optima as Funnels in Fitness Landscapes

2016

We conduct an analysis of local optima networks extracted from fitness landscapes of the Kauffman NK model under iterated local search. Applying the Markov Cluster Algorithm for community detection to the local optima networks, we find that the landscapes consist of multiple clusters. This result complements recent findings in the literature that landscapes often decompose into multiple funnels, which increases their difficulty for iterated local search. Our results suggest that the number of clusters as well as the size of the cluster in which the global optimum is located are correlated to the search difficulty of landscapes. We conclude that clusters found by community detection in local…

Bayesian estimation of edge orientations in junctions

1999

Abstract Junctions, defined as those points of an image where two or more edges meet, play a significant role in many computer vision applications. Junction detection is a widely treated problem, and some detectors can provide even the directions of the edges that meet in a junction. The main objective of this paper is the precise estimation of such directions. It is supposed that the junction point has been previously found by some detector. Also, it is assumed that samples, possibly noisy, of orientations of the edges found in a circular window surrounding the point are available. A mixture of von Mises distributions is assumed for these data, and then a Bayesian methodology is applied to…

A new strategy for effective learning in population Monte Carlo sampling

2016

In this work, we focus on advancing the theory and practice of a class of Monte Carlo methods, population Monte Carlo (PMC) sampling, for dealing with inference problems with static parameters. We devise a new method for efficient adaptive learning from past samples and weights to construct improved proposal functions. It is based on assuming that, at each iteration, there is an intermediate target and that this target is gradually getting closer to the true one. Computer simulations show and confirm the improvement of the proposed strategy compared to the traditional PMC method on a simple considered scenario.

Bayesian adaptive estimation: The next dimension

2006

Abstract We propose a new psychometric model for two-dimensional stimuli, such as color differences, based on parameterizing the threshold of a one-dimensional psychometric function as an ellipse. The Ψ Bayesian adaptive estimation method applied to this model yields trials that vary in multiple stimulus dimensions simultaneously. Simulations indicate that this new procedure can be much more efficient than the more conventional procedure of estimating the psychometric function on one-dimensional lines independently, requiring only one-fourth or less the number of trials for equivalent performance in typical situations. In a real psychophysical experiment with a yes–no task, as few as 22 tri…

Hydrological post-processing based on approximate Bayesian computation (ABC)

2019

[EN] This study introduces a method to quantify the conditional predictive uncertainty in hydrological post-processing contexts when it is cumbersome to calculate the likelihood (intractable likelihood). Sometimes, it can be difficult to calculate the likelihood itself in hydrological modelling, specially working with complex models or with ungauged catchments. Therefore, we propose the ABC post-processor that exchanges the requirement of calculating the likelihood function by the use of some sufficient summary statistics and synthetic datasets. The aim is to show that the conditional predictive distribution is qualitatively similar produced by the exact predictive (MCMC post-processor) or …

Cross-entropy-based adaptive optimization of simulation parameters for Markovian-driven service systems

2005

Abstract Markov fluid models represent a general description of the process of service request arrivals to service systems. The solution of performance analysis problems incorporating them often calls for a simulation approach, for which a reference methodology is Importance Sampling. However, in this case the appropriate choice of the biasing conditions is a problem in itself. In this paper an iterative method based on the cross-entropy is proposed for this choice. The equations are given that allow to derive the biasing conditions from the simulation itself. The application of the proposed method to three different sample cases, referring to one transient scenario (finite time horizon and…

Hydropower Optimization Using Deep Learning

2019

This paper demonstrates how deep learning can be used to find optimal reservoir operating policies in hydropower river systems. The method that we propose is based on the implicit stochastic optimization (ISO) framework, using direct policy search methods combined with deep neural networks (DNN). The findings from a real-world two-reservoir hydropower system in southern Norway suggest that DNNs can learn how to map input (price, inflow, starting reservoir levels) to the optimal production pattern directly. Due to the speed of evaluating the DNN, this approach is from an operational standpoint computationally inexpensive and may potentially address the long-standing problem of high dimension…