New results of Quick Learning for Bidirectional Associative Memory having high capacity

Author

Hattori, Motonobu ; Hagiwara, Masafumi ; Nakagawa, Masao

Author_Institution

Dept. of Electr. Eng., Keio Univ., Yokohama, Japan

Volume

2

fYear

1994

fDate

27 Jun-2 Jul 1994

Firstpage

1080

Abstract

Several important characteristics of Quick Learning for Bidirectional Associative Memory (QLBAM) are introduced. QLBAM uses two stage learning. In the first stage, the BAM is trained by Hebbian learning and then by Pseudo-Relaxation Learning Algorithm for BAM (PRLAB). The following features of the QLBAM are made clear: it is insensitive to correlation of training pairs; it is robust for noisy inputs; the minimum absolute value of net inputs indexes a noise margin; the memory capacity is greatly improved: the maximum capacity in our simulation is about 2.2N

Keywords

Hebbian learning; content-addressable storage; learning (artificial intelligence); neural nets; Hebbian learning; PRLAB; Pseudo-Relaxation Learning Algorithm; QLBAM; Quick Learning for Bidirectional Associative Memory; correlation; high capacity; maximum capacity; memory capacity; minimum absolute value; net inputs; noise margin; noisy inputs; robust; simulation; training pairs; two stage learning; Associative memory; Biological neural networks; Brain modeling; Hebbian theory; Magnesium compounds; Neurons; Noise reduction; Noise robustness; Relaxation methods; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on

Conference_Location

Orlando, FL

Print_ISBN

0-7803-1901-X

Type

conf

DOI

10.1109/ICNN.1994.374333

Filename

374333