A MULTILAYER NEURAL NETWORK WITH PIECEWISE-LINEAR STRUCTURE AND BACK-PROPAGATION LEARNING

被引：32

作者：

BATRUNI, R

机构：

[1] National Semiconductor Corporation, Santa Clara, CA 95051, MS 11–125

来源：

IEEE TRANSACTIONS ON NEURAL NETWORKS | 1991年 / 2卷 / 03期

关键词：

D O I：

10.1109/72.97915

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper proposes a new multilayer neural network which is given a two-layer piecewise-linear structure for every cascaded section. Steepest descent and back-propagation perform convergence of the network. This new form of neural networks specializes in functional approximation and is anticipated to have applications in control, communications, and pattern recognition.

引用

页码：395 / 403

页数：9

共 22 条

[1]

[Anonymous], 1989, ANALOG VLSI NEURAL S

[2] ESCAPE BEHAVIOR OF THE COCKROACH PERIPLANETA-AMERICANA .2. DETECTION OF NATURAL PREDATORS BY AIR DISPLACEMENT [J].

CAMHI, JM ;

TOM, W ;

VOLMAN, S .

JOURNAL OF COMPARATIVE PHYSIOLOGY, 1978, 128 (03) :203-212

[3]

CAMHI JM, 1978, J COMP PHYSIOL, V128, P193, DOI 10.1007/BF00656852

[4]

CHUA L, 1988, IEEE T CIRC SYST, V35

[5]

CHUA LO, 1986, IEEE T CIRC SYST, V33

[6]

DEMBO A, 1990, IEEE T NEURAL NETWOR, V1

[7] MULTIVARIATE STATISTICAL-ANALYSIS OF THE RESPONSE OF THE COCKROACH GIANT INTERNEURON SYSTEM TO WIND PUFFS [J].

HEETDERKS, WJ ;

BATRUNI, R .

BIOLOGICAL CYBERNETICS, 1982, 43 (01) :1-11

[8]

HUTCHINSON J, 1988, COMPUTER MAR

[9]

KAHLERT C, 1990, IEEE T CIRC SYST, V37

[10]

Kangas Jari A., 1990, IEEE T NEURAL NETWOR, V1

← 1 2 3 →