L1 NORMALIZATION IN NEURAL NETWORKS USING PARALLEL AUTOMATIC LEVEL CONTROL

被引：1

作者：

PERFETTI, R

机构：

[1] Info-Com Dpt., Universita di Roma ‘La Sapienza’, 00184 Roma

来源：

ELECTRONICS LETTERS | 1990年 / 26卷 / 02期

关键词：

Artificial intelligence; Circuit theory and design; Networks; Neural networks; Pattern recognition;

D O I：

10.1049/el:19900055

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

A simple circuit that performs L1 normalisation of positivevalued parallel inputs is proposed. It can be utilised to implement ‘on-line’ normalisation of input patterns or weight vectors in neural networks based on competitive learning mechanisms. A modified scheme with high linearity is also presented. © 1990, The Institution of Electrical Engineers. All rights reserved.

引用

页码：82 / 84

页数：3

共 6 条

[1] GRAF HP, 1989, IEEE CIRC DEV MAG, V5
[2] ADAPTIVE PATTERN-CLASSIFICATION AND UNIVERSAL RECODING .1. PARALLEL DEVELOPMENT AND CODING OF NEURAL FEATURE DETECTORS
GROSSBERG, S
[J]. BIOLOGICAL CYBERNETICS, 1976, 23 (03) : 121 - 134
[3] Kohonen T., 1987, SELF ORG ASS MEMORY
[4] Rumelhart DE, 1986, ENCY DATABASE SYST, P45
[5] CONTINUOUS-TIME MOSFET-C FILTERS IN VLSI
TSIVIDIS, Y
BANU, M
KHOURY, J
[J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS, 1986, 33 (02): : 125 - 140
[6] TSIVIDIS Y, 1986, OPERATION MODELING M

← 1 →