Education, Science, Technology, Innovation and Life
Open Access
Sign In

The Research of Deep Belief Network

Download as PDF

DOI: 10.23977/iset.2019.013

Author(s)

Zizhao Chen, Wencheng Jiao

Corresponding Author

Zizhao Chen

ABSTRACT

As a hot research field in recent years, deep learning has great application prospects. First the technical research results and the application fields of deep learning is introduced in this paper. Second the DBN network model is taken as an example to introduce its theoretical basis, network structure and training process. Then we analyze the research thoughts and achievements of experts and scholars in recent years and propose to classify the main research directions into three categories, containing DBN input data, DBN network structure and DBN parameter optimization. At the same time, the literature in various fields at home and abroad in recent years are classified according to the three categories mentioned. Finally, the research trends of the deep belief network are analyzed, in order to provide researchers with ideas for improving DBN.

KEYWORDS

Deep Learning, DBN, RBM, input data, network structure, parameter optimization

All published work is licensed under a Creative Commons Attribution 4.0 International License.

Copyright © 2016 - 2031 Clausius Scientific Press Inc. All Rights Reserved.