Enregistré dans:
Détails bibliographiques
Auteurs principaux: Richardson, Oliver E., Samiei, Mandana, Shakerinava, Mehran, Viviano, Joseph D., Kabid, Abdessamad El, Parviz, Ali, Bengio, Yoshua
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2604.17140
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918453311438848
author Richardson, Oliver E.
Samiei, Mandana
Shakerinava, Mehran
Viviano, Joseph D.
Kabid, Abdessamad El
Parviz, Ali
Bengio, Yoshua
author_facet Richardson, Oliver E.
Samiei, Mandana
Shakerinava, Mehran
Viviano, Joseph D.
Kabid, Abdessamad El
Parviz, Ali
Bengio, Yoshua
contents We present a generic algorithm for learning and approximate inference with an intuitive epistemic interpretation: iteratively focus on a subset of the model and resolve inconsistencies using the parameters under control. This framework, which we call Local Inconsistency Resolution (LIR) is built upon Probabilistic Dependency Graphs (PDGs), which provide a flexible representational foundation capable of capturing inconsistent beliefs. We show how LIR unifies and generalizes a wide variety of important algorithms in the literature, including the Expectation-Maximization (EM) algorithm, belief propagation, adversarial training, GANs, and GFlowNets. In the last case, LIR actually suggests a more natural loss, which we demonstrate improves GFlowNet convergence. Each method can be recovered as a specific instance of LIR by choosing a procedure to direct focus (attention and control). We implement this algorithm for discrete PDGs and study its properties on synthetically generated PDGs, comparing its behavior to the global optimization semantics of the full PDG.
format Preprint
id arxiv_https___arxiv_org_abs_2604_17140
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Local Inconsistency Resolution: The Interplay between Attention and Control in Probabilistic Models
Richardson, Oliver E.
Samiei, Mandana
Shakerinava, Mehran
Viviano, Joseph D.
Kabid, Abdessamad El
Parviz, Ali
Bengio, Yoshua
Artificial Intelligence
Machine Learning
We present a generic algorithm for learning and approximate inference with an intuitive epistemic interpretation: iteratively focus on a subset of the model and resolve inconsistencies using the parameters under control. This framework, which we call Local Inconsistency Resolution (LIR) is built upon Probabilistic Dependency Graphs (PDGs), which provide a flexible representational foundation capable of capturing inconsistent beliefs. We show how LIR unifies and generalizes a wide variety of important algorithms in the literature, including the Expectation-Maximization (EM) algorithm, belief propagation, adversarial training, GANs, and GFlowNets. In the last case, LIR actually suggests a more natural loss, which we demonstrate improves GFlowNet convergence. Each method can be recovered as a specific instance of LIR by choosing a procedure to direct focus (attention and control). We implement this algorithm for discrete PDGs and study its properties on synthetically generated PDGs, comparing its behavior to the global optimization semantics of the full PDG.
title Local Inconsistency Resolution: The Interplay between Attention and Control in Probabilistic Models
topic Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2604.17140