• Home
  • Search
  • Approximate counting, the Lovasz local lemma, and inference in graphical models
  • Cite Icon19
  • https://doi.org/10.1145/3055399.3055428Copy DOI Icon

Approximate counting, the Lovasz local lemma, and inference in graphical models

  • Jun 19, 2017
  • Ankur Moitra
Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

In this paper we introduce a new approach for approximately counting in bounded degree systems with higher-order constraints. Our main result is an algorithm to approximately count the number of solutions to a CNF formula Ф when the width is logarithmic in the maximum degree. This closes an exponential gap between the known upper and lower bounds. Moreover our algorithm extends straightforwardly to approximate sampling, which shows that under Lovasz Local Lemma-like conditions it is not only possible to find a satisfying assignment, it is also possible to generate one approximately uniformly at random from the set of all satisfying assignments. Our approach is a significant departure from earlier techniques in approximate counting, and is based on a framework to bootstrap an oracle for computing marginal probabilities on individual variables. Finally, we give an application of our results to show that it is algorithmically possible to sample from the posterior distribution in an interesting class of graphical models.

Similar Papers
  • Supplementary Content
  • Citations8

Approximate Model Counting: Is SAT Oracle More Powerful than NP Oracle?

  • Jun 17, 2023
  • arXiv (Cornell University)
  • Diptarka Chakraborty +3
  • Single Report
  • Citations5

Statistical Inference in Graphical Models

  • Jun 17, 2008
  • K Gimpel +1
  • Conference Article
  • Citations2

Local Approximations of the Independent Set Polynomial

  • Jan 01, 2021
  • DROPS (Schloss Dagstuhl – Leibniz Center for Informatics)
  • Dimitris Achlioptas +1
  • Research Article
  • Citations17

Improving probabilistic inference in graphical models with determinism and cycles

  • Jul 22, 2016
  • Machine Learning
  • Mohamed-Hamza Ibrahim +2
  • Research Article
  • Citations4

Parallel Algorithms for GPU accelerated Probabilistic Inference

  • Feb 21, 2012
  • Technische Universität Dortmund Eldorado (Technische Universität Dortmund)
  • Nico Piatkowski
  • Research Article
  • Citations44

Acyclic edge coloring through the Lovász Local Lemma

  • Dec 23, 2016
  • Theoretical Computer Science
  • Ioannis Giotis +3
  • Research Article
  • Citations4

Learning Gaussian graphical models with latent confounders

  • Jul 13, 2023
  • Journal of Multivariate Analysis
  • Ke Wang +2
  • Research Article
  • Citations9

The Linearization of Belief Propagation on Pairwise Markov Random Fields

  • Feb 12, 2017
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Wolfgang Gatterbauer
  • Book Chapter
  • Citations109

A Bayesian model of imitation in infants and robots

  • Mar 29, 2007
  • Rajesh P N Rao +2
  • Book Chapter
  • Citations7

On the Impact of Junction-Tree Topology on Weighted Model Counting

  • Jan 01, 2015
  • Batya Kenig +1
  • Supplementary Content
  • Citations2

Planning and exploration in stochastic relational worlds

  • Jan 01, 2011
  • Universitätsbibliothek der FU Berlin Hochschulschriftenstelle u. Dokumentenserver
  • Tobias Lang
  • Research Article
  • Citations3

Stopping Rules for Randomized Greedy Triangulation Schemes

  • Aug 04, 2011
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Andrew Gelfand +2
  • Conference Article
  • Citations1

A scale-free structure prior for Bayesian inference of Gaussian graphical models

  • Nov 01, 2014
  • Osamu Maruyama +1
  • Research Article
  • Citations32

Lifted Probabilistic Inference for Asymmetric Graphical Models

  • Mar 04, 2015
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Guy Van Den Broeck +1
  • Supplementary Content

Query-specific learning and inference for probabilistic graphical models

  • Jan 01, 2011
  • Figshare
  • Carlos Guestrin +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.