Algorithms for estimating the partition function of restricted Boltzmann machines

DOI10.1016/J.ARTINT.2019.103195zbMATH Open1478.68291OpenAlexW2982859738WikidataQ78314631 ScholiaQ78314631MaRDI QIDQ2287191FDOQ2287191

Authors: Oswin Krause, Asja Fischer, Christian Igel 

Publication date: 20 January 2020

Published in: Artificial Intelligence (Search for Journal in Brave)

Full work available at URL: https://doi.org/10.1016/j.artint.2019.103195

Recommendations

• Stochastic approximation algorithms for partition function estimation of Gibbs random fields
• A Novel Parameter Estimation Method for Boltzmann Machines
• A bound for the convergence rate of parallel tempering for sampling restricted Boltzmann machines
• Approximation algorithms for the normalizing constant of Gibbs distributions
• The flip-the-state transition operator for restricted Boltzmann machines

zbMATH Keywords

parallel temperingIsing modelrestricted Boltzmann machinesbridge samplingannealed importance samplingBennett's acceptance ratioCrooks' equalitypartition function estimation

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Probabilistic graphical models (62H22) Lattice systems (Ising, dimer, Potts, etc.) and systems on graphs arising in equilibrium statistical mechanics (82B20)

Cites Work

• \texttt{SHARK}
• Title not available (Why is that?)
• Simulating normalizing constants: From importance sampling to bridge sampling to path sampling
• Beitrag zur Theorie des Ferromagnetismus
• Training Products of Experts by Minimizing Contrastive Divergence
• The flip-the-state transition operator for restricted Boltzmann machines
• A bound for the convergence rate of parallel tempering for sampling restricted Boltzmann machines
• Training restricted Boltzmann machines: an introduction

Cited In (10)

• A Novel Parameter Estimation Method for Boltzmann Machines
• Estimating a Potential Without the Agony of the Partition Function
• Learning a restricted Boltzmann machine using biased Monte Carlo sampling
• Stochastic approximation algorithms for partition function estimation of Gibbs random fields
• Inferring effective couplings with restricted Boltzmann machines
• Tradeoff of generalization error in unsupervised learning
• Equilibrium and non-equilibrium regimes in the learning of restricted Boltzmann machines*
• Deep learning-based approximation of Goldbach partition function
• Replica analysis of the lattice-gas restricted Boltzmann machine partition function
• Title not available (Why is that?)

Uses Software

• SHARK

This page was built for publication: Algorithms for estimating the partition function of restricted Boltzmann machines

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2287191)