An important part of problems in statistical physics and computer science can be expressed as the computation of marginal probabilities over a Markov Random Field. The belief propagation algorithm, which is an exact procedure to compute these marginals when the underlying graph is a tree, has gained its popularity as an efficient way to approximate them in the more general case. In this paper, we focus on an aspect of the algorithm that did not get that much attention in the literature, which is the effect of the normalization of the messages. We show in particular that, for a large class of normalization strategies, it is possible to focus only on belief convergence. Following this, we express the necessary and sufficient conditions for lo...
According to physics predictions, the free energy of random factor graph models that satisfy a certa...
We replace strong independence in credal networks with the weaker notion of epistemic irrelevance. F...
In this work, we focus on the design and estimation - from partial observations - of graphical model...
An important part of problems in statistical physics and computer science can be expressed as the co...
International audienceA number of problems in statistical physics and computer science can be expres...
Les algorithmes à propagation de messages constituent un schéma de calcul parallèle pour estimer les...
International audienceLarge scale inference problems of practical interest can often be addressed wi...
On s'intéresse à la construction et l'estimation - à partir d'observations incomplètes - de modèles ...
Dans cette thèse, nous étudions le problème de l'inférence bayésienne dans les graphes factoriels, e...
Important inference problems in statistical physics, computer vision, error-correcting coding theory...
Belief propagation (BP) was only supposed to work for tree-like networks but works surprisingly well...
Contains fulltext : 72395.pdf (publisher's version ) (Open Access)The research rep...
Loopy belief propagation performs approximate inference on graphical models with loops. One might ho...
Abstract—Inference problems in graphical models can be rep-resented as a constrained optimization of...
When belief propagation (BP) converges, it does so to a stationary point of the Bethe free energy F,...
According to physics predictions, the free energy of random factor graph models that satisfy a certa...
We replace strong independence in credal networks with the weaker notion of epistemic irrelevance. F...
In this work, we focus on the design and estimation - from partial observations - of graphical model...
An important part of problems in statistical physics and computer science can be expressed as the co...
International audienceA number of problems in statistical physics and computer science can be expres...
Les algorithmes à propagation de messages constituent un schéma de calcul parallèle pour estimer les...
International audienceLarge scale inference problems of practical interest can often be addressed wi...
On s'intéresse à la construction et l'estimation - à partir d'observations incomplètes - de modèles ...
Dans cette thèse, nous étudions le problème de l'inférence bayésienne dans les graphes factoriels, e...
Important inference problems in statistical physics, computer vision, error-correcting coding theory...
Belief propagation (BP) was only supposed to work for tree-like networks but works surprisingly well...
Contains fulltext : 72395.pdf (publisher's version ) (Open Access)The research rep...
Loopy belief propagation performs approximate inference on graphical models with loops. One might ho...
Abstract—Inference problems in graphical models can be rep-resented as a constrained optimization of...
When belief propagation (BP) converges, it does so to a stationary point of the Bethe free energy F,...
According to physics predictions, the free energy of random factor graph models that satisfy a certa...
We replace strong independence in credal networks with the weaker notion of epistemic irrelevance. F...
In this work, we focus on the design and estimation - from partial observations - of graphical model...