Fast convergence with low precision weights in ART1 networks

Author

Crespo, Jean-francois ; Lavoie, Pierre ; Savaria, Yvon

Author_Institution

Dept. of Electr. & Comput. Eng., Ecole Polytech. de Montreal, Que., Canada

Volume

6

fYear

1994

fDate

30 May-2 Jun 1994

Firstpage

237

Abstract

A new learning law, the Direct Coding Rule, is proposed for bottom-up long term memory learning in Adaptive Resonance Theory (ART) networks. This law requires less computational precision than the traditional Weber Law Rule and modifies the search dynamics of the network to accelerate convergence. Following a brief mathematical analysis of the new learning law, an ART1 network based on this law is applied to a passive radar detection problem. The simulation results allow comparison of the new law to the Weber Law Rule, with and without weight quantization, from the speed and cost viewpoints

Keywords

ART neural nets; convergence; encoding; learning (artificial intelligence); ART1 networks; adaptive resonance theory networks; bottom-up long term memory learning; direct coding rule; fast convergence; learning law; low precision weights; passive radar detection problem; weight quantization; Acceleration; Computational modeling; Computer networks; Convergence; Mathematical analysis; Passive radar; Quantization; Radar detection; Resonance; Subspace constraints;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 1994. ISCAS '94., 1994 IEEE International Symposium on

Conference_Location

London

Print_ISBN

0-7803-1915-X

Type

conf

DOI

10.1109/ISCAS.1994.409571

Filename

409571