Presentation 1998/11/17
A neural network model with excitable map and its application to temporal learning
Yoshinori Hayakawa, Yasuji Sawada,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose a pulse-coupled neural network model in which one-dimensional excitable map is introduced as the neural processing unit connected with time delay. Although individual processing unit solely has simple dynamical properties, the network behaves as a collective chaotic system with robust synchronization within a portion of the units. By introducing a covariance learning algorithm for synaptic connections, we find that the synchronization of excitation timing of the unit can be enhanced and such synchronization can be even controlled by external signals. The learnable range of time is much longer than the inherent time scale of the processing unit, i.e., synaptic delay time. Also, the network can reproduce periodic signals in higher resolution of time than such characteristic time scale.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) one-dimensional map / neural network / temporal learning / synchronization
Paper # NC98-61
Date of Issue

Conference Information
Committee NC
Conference Date 1998/11/17(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Neurocomputing (NC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A neural network model with excitable map and its application to temporal learning
Sub Title (in English)
Keyword(1) one-dimensional map
Keyword(2) neural network
Keyword(3) temporal learning
Keyword(4) synchronization
1st Author's Name Yoshinori Hayakawa
1st Author's Affiliation Research Institute of Electrical Communication, Tohoku university()
2nd Author's Name Yasuji Sawada
2nd Author's Affiliation Research Institute of Electrical Communication, Tohoku university
Date 1998/11/17
Paper # NC98-61
Volume (vol) vol.98
Number (no) 401
Page pp.pp.-
#Pages 8
Date of Issue