An Adaptive Learning Method of Deep Belief Network by Layer Generation Algorithm

TL;DR

This paper introduces an adaptive learning method for Deep Belief Networks that dynamically determines the optimal number of layers and neurons during training, improving efficiency and model stability.

Contribution

It presents a novel adaptive learning algorithm for DBNs that automatically adjusts network structure during training, unlike traditional fixed-structure models.

Findings

Successfully determined optimal layers on benchmark datasets

Improved computational efficiency and model stability

Enhanced feature representation through adaptive structure

Abstract

Deep Belief Network (DBN) has a deep architecture that represents multiple features of input patterns hierarchically with the pre-trained Restricted Boltzmann Machines (RBM). A traditional RBM or DBN model cannot change its network structure during the learning phase. Our proposed adaptive learning method can discover the optimal number of hidden neurons and weights and/or layers according to the input space. The model is an important method to take account of the computational cost and the model stability. The regularities to hold the sparse structure of network is considerable problem, since the extraction of explicit knowledge from the trained network should be required. In our previous research, we have developed the hybrid method of adaptive structural learning method of RBM and Learning Forgetting method to the trained RBM. In this paper, we propose the adaptive learning method of DBN that can determine the optimal number of layers during the learning. We evaluated our proposed model on some benchmark data sets.