Equilibrium and non-Equilibrium regimes in the learning of Restricted Boltzmann Machines.

Aurélien Decelle Cyril Furtlehner Beatriz Seoane

Published in: NeurIPS (2021)

Keyphrases

learning algorithm
learning process
reinforcement learning
supervised learning
deep learning
maximum likelihood
neural network
information retrieval
feature selection
active learning
unsupervised learning
restricted boltzmann machine
deep belief networks
contrastive divergence