An Adaptive Learning Method of Restricted Boltzmann Machine by Neuron Generation and Annihilation Algorithm

TL;DR

This paper introduces an adaptive learning method for Restricted Boltzmann Machines that dynamically adjusts the number of hidden neurons during training using neuron generation and annihilation algorithms, improving structure optimization.

Contribution

It proposes a novel adaptive RBM training approach that automatically determines the optimal hidden neuron count based on parameter variance and energy function convergence.

Findings

Effective hidden neuron adjustment during training

Improved model performance on benchmark datasets

Reduced need for manual network structure tuning

Abstract

Restricted Boltzmann Machine (RBM) is a generative stochastic energy-based model of artificial neural network for unsupervised learning. Recently, RBM is well known to be a pre-training method of Deep Learning. In addition to visible and hidden neurons, the structure of RBM has a number of parameters such as the weights between neurons and the coefficients for them. Therefore, we may meet some difficulties to determine an optimal network structure to analyze big data. In order to evade the problem, we investigated the variance of parameters to find an optimal structure during learning. For the reason, we should check the variance of parameters to cause the fluctuation for energy function in RBM model. In this paper, we propose the adaptive learning method of RBM that can discover an optimal number of hidden neurons according to the training situation by applying the neuron generation and annihilation algorithm. In this method, a new hidden neuron is generated if the energy function is not still converged and the variance of the parameters is large. Moreover, the inactivated hidden neuron will be annihilated if the neuron does not affect the learning situation. The experimental results for some benchmark data sets were discussed in this paper.