A LEARNING-PROCESS OF A STOCHASTIC FEEDFORWARD NEURAL-NETWORK

被引：2

作者：

FUJIKI, S ^{[1
]}

FUJIKI, NM ^{[1
]}

机构：

[1] SENDAI NATL COLL TECHNOL, SENDAI, MIYAGI 98931, JAPAN

来源：

JOURNAL OF THE PHYSICAL SOCIETY OF JAPAN | 1995年 / 64卷 / 03期

关键词：

A LEARNING PROCESS; FEEDFORWARD NEURAL NETWORK; STOCHASTIC LAYERED NETWORK;

D O I：

10.1143/JPSJ.64.757

中图分类号：

O4 [物理学];

学科分类号：

0702 ;

摘要：

A positive reinforcement type learning algorithm is formulated for a stochastic feed-forward neural network by minimizing a relative entropic measure, and a learning equation similar to that of the Boltzmann machine is obtained. The learning of the network actually shows a similar result to that of the Boltzmann machine in the classification problems of AND and XOR, by numerical experiments.

引用

页码：757 / 765

页数：9

共 50 条

[1] THE LEARNING-PROCESS IN BIOFEEDBACK - IS IT FEEDFORWARD OR FEEDBACK
DUNN, T
GILLIG, S
PONSOR, S
WEIL, N
UTZ, SW
BIOFEEDBACK AND SELF-REGULATION, 1985, 10 (01): : 87 - 87
[2] THE LEARNING-PROCESS IN BIOFEEDBACK - IS IT FEEDFORWARD OR FEEDBACK
DUNN, TG
GILLIG, SE
PONSOR, SE
WEIL, N
UTZ, SW
BIOFEEDBACK AND SELF-REGULATION, 1986, 11 (02): : 143 - 156
[3] Biomimetic Hybrid Feedback Feedforward Neural-Network Learning Control
Pan, Yongping
Yu, Haoyong
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2017, 28 (06) : 1481 - 1487
[4] CONSULTING AS A STOCHASTIC LEARNING-PROCESS
MENOU, MJ
PROCEEDINGS OF THE AMERICAN SOCIETY FOR INFORMATION SCIENCE, 1982, 19 : 372 - 372
[5] Moving-window spectral neural-network feedforward process control
Ridley, D
Llaugel, F
IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, 2000, 47 (03) : 393 - 402
[6] STOCHASTIC SENSITIVITY ANALYSIS METHOD FOR NEURAL-NETWORK LEARNING
KODA, M
INTERNATIONAL JOURNAL OF SYSTEMS SCIENCE, 1995, 26 (03) : 703 - 711
[7] ON THE GEOMETRY OF FEEDFORWARD NEURAL-NETWORK ERROR SURFACES
CHEN, AM
LU, HM
HECHTNIELSEN, R
NEURAL COMPUTATION, 1993, 5 (06) : 910 - 927
[8] RECURRENT NEURAL-NETWORK TRAINING WITH FEEDFORWARD COMPLEXITY
OLUROTIMI, O
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1994, 5 (02): : 185 - 197
[9] ELECTRONIC IMPLEMENTATION OF AN ANALOG ATTRACTOR NEURAL-NETWORK WITH STOCHASTIC LEARNING
BADONI, D
BERTAZZONI, S
BUGLIONI, S
SALINA, G
AMIT, DJ
FUSI, S
NETWORK-COMPUTATION IN NEURAL SYSTEMS, 1995, 6 (02) : 125 - 157
[10] Direct Learning for Parameter-Varying Feedforward Control: A Neural-Network Approach
Kon, Johan
van de Wijdeven, Jeroen
Bruijnen, Dennis
Toth, Roland
Heertjes, Marcel
Oomen, Tom
2023 62ND IEEE CONFERENCE ON DECISION AND CONTROL, CDC, 2023, : 3720 - 3725

← 1 2 3 4 5 →