Learning Coefficient in Bayesian Estimation of Restricted Boltzmann Machine

Abstract

We consider the real log canonical threshold for the learning model in Bayesian estimation.This threshold corresponds to a learning coefficient of generalization error in Bayesianestimation, which serves to measure learning efficiency in hierarchical learning models [30, 31, 33].In this paper, we clarify the ideal which gives the log canonical threshold of the restricted Boltzmannmachine and consider the learning coefficients of this model.