STRUCTURE OPTIMIZATION OF DEEP BELIEF NETS IN THE APPLICATIONS OF IMAGE RECOGNITION
Journal: International Journal of Engineering Sciences & Research Technology (IJESRT) (Vol.9, No. 7)Publication Date: 2020-07-30
Authors : Qili CHEN Guangyuan PAN; Ming Yu;
Page : 179-189
Keywords : Deep learning; Structure analysis; Unsupervised learning; Image recognition.;
Abstract
Deep belief network (DBN) has become one of the most important models in deep learning, however, the unoptimized structure leads to wasting too much training resources. To solve this problem and to investigate the connection of depth and accuracy of DBN, an optimization training method that consists of two steps is proposed. Firstly, by using mathematical and biological tools, the significance of supervised training is analyzed, and a theorem, that is on reconstruction error and network energy, is proved. Secondly, based on conclusions of step one, this paper proposes to optimize the structure of DBN (especially hidden layer numbers). Thirdly, this method is applied in two image recognition experiments, and results show increased computing efficiency and accuracies in both tasks.
Other Latest Articles
- IGNATOV ALEXEY NIKOLAEVICH AND HIS SCIENTIFIC HERITAGE
- IN MEMORY OF HONORED SCIENTIST OF THE RUSSIAN FEDERATION, DOCTOR OF LAW PROF. V.K. PUCHINSKY
- DEPARTMENT OF FORENSIC ACTIVITY IS AN EXAMPLE OF THE SYNTHESIS OF SCIENCE AND FORENSIC PRACTICES AND EDUCATION
- PROBLEMS OF THE FIGHT AGAINST CRIME IN THE XXI CENTURY: CRIMINAL LAW, CRIMINAL PROCEDURE AND CRIMINALISTICS DECISIONS
- SCIENTIFIC SCHOOL OF PRIVATE LAW OF RESEARCH AT THE PEOPLES' FRIENDSHIP UNIVERSITY OF RUSSIA: DEPARTMENT OF CIVIL LAW AND PROCESS AND PRIVATE INTERNATIONAL LAW
Last modified: 2020-08-30 06:49:40