Presentation 1994/10/13
A Biologically Plausible Neural Network (BPNN) Model for Recognition and Generation of Temporal Sequences
Ryoko Futami, Nozomu Hoshimiya,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) A neural network model for the recognition and generation of temporal sequences is proposed.This model is composed of Kohonen's self-organizing feature maps and the feedback through delay or memory units which have smooth temporal responses.It is shown by computer simulation that this model can be self-organized to satisfy desired input-output relationships.As this model uses feature maps,some advantages of local representation,i.e.,fast learning speed and robustness for the additional learning on new sequences can be expected.Those features have not been realized by the combination of recurrent-type neural network model and error back-propagation learning.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Temporal Sequence / Local representation / Feature map / Recognition / Generation / Imitation
Paper # NC94-39
Date of Issue

Conference Information
Committee NC
Conference Date 1994/10/13(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 Biologically Plausible Neural Network (BPNN) Model for Recognition and Generation of Temporal Sequences
Sub Title (in English)
Keyword(1) Temporal Sequence
Keyword(2) Local representation
Keyword(3) Feature map
Keyword(4) Recognition
Keyword(5) Generation
Keyword(6) Imitation
1st Author's Name Ryoko Futami
1st Author's Affiliation Faculty of Engineering,Tohoku University()
2nd Author's Name Nozomu Hoshimiya
2nd Author's Affiliation Faculty of Engineering,Tohoku University
Date 1994/10/13
Paper # NC94-39
Volume (vol) vol.94
Number (no) 272
Page pp.pp.-
#Pages 8
Date of Issue