Latent binary MRF for online reconstruction of large scale systems

DOI10.1007/S10472-015-9470-XzbMATH Open1401.68266arXiv1312.6607OpenAlexW3099846819MaRDI QIDQ505979FDOQ505979

Authors: Victorin Martin, Jean-Marc Lasgouttes, Cyril Furtlehner 

Publication date: 27 January 2017

Published in: Annals of Mathematics and Artificial Intelligence (Search for Journal in Brave)

Abstract: We propose a probabilistic graphical model realizing a minimal encoding of real variables dependencies based on possibly incomplete observation and an empirical cumulative distribution function per variable. The target application is a large scale partially observed system, like e.g. a traffic network, where a small proportion of real valued variables are observed, and the other variables have to be predicted. Our design objective is therefore to have good scalability in a real-time setting. Instead of attempting to encode the dependencies of the system directly in the description space, we propose a way to encode them in a latent space of binary variables, reflecting a rough perception of the observable (congested/non-congested for a traffic road). The method relies in part on message passing algorithms, i.e. belief propagation, but the core of the work concerns the definition of meaningful latent variables associated to the variables of interest and their pairwise dependencies. Numerical experiments demonstrate the applicability of the method in practice.

Full work available at URL: https://arxiv.org/abs/1312.6607

Recommendations

• An Ising model for road traffic inference
• Network tomography for integer-valued traffic
• Pairwise MRF models selection for traffic inference
• Propagation of information on undirected dependency graphs for road traffic inference
• Traffic data reconstruction based on Markov random field modeling

zbMATH Keywords

latent variablesinferenceMarkov random fieldbelief propagationsoft constraints

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Random fields (60G60) Interacting random processes; statistical mechanics type models; percolation theory (60K35)

Cites Work

• The PUMAS library
• Title not available (Why is that?)
• Nearest neighbor pattern classification
• Sparse inverse covariance estimation with the graphical lasso
• Title not available (Why is that?)
• High-dimensional Ising model selection using \(\ell _{1}\)-regularized logistic regression
• Title not available (Why is that?)
• Title not available (Why is that?)
• Probability Theory
• Generalized Iterative Scaling for Log-Linear Models
• Constructing Free-Energy Approximations and Generalized Belief Propagation Algorithms
• Prior Probabilities
• Adaptive cluster expansion for the inverse Ising problem: convergence, algorithm and tests
• Factor graphs and the sum-product algorithm
• Title not available (Why is that?)
• Approximate inference in Boltzmann machines
• Approximate inverse Ising models close to a Bethe reference point
• Approximate Learning Algorithm in Boltzmann Machines
• On the revision of probabilistic beliefs using uncertain evidence
• Loopy belief propagation: convergence and effects of message errors
• Estimating the ``wrong graphical model: benefits in the computation-limited setting
• Sufficient Conditions for Convergence of the Sum–Product Algorithm

Cited In (5)

• Pairwise MRF models selection for traffic inference
• An Ising model for road traffic inference
• Cycle-based cluster variational method for direct and inverse inference
• Propagation of information on undirected dependency graphs for road traffic inference
• Traffic data reconstruction based on Markov random field modeling

Uses Software

• glasso

This page was built for publication: Latent binary MRF for online reconstruction of large scale systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q505979)