Presentation 2001/3/16
An Incremental Learning and Reduction of Redundant Cells by Sleeping
Takahiro TAGAWA, Koichiro YAMAUCHI, Naohiro ISHII, Sachio ITO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) One advantage of artificial neural networks is the function approximation with a small number of units. However, the neural consisted of small number of units needs a large number of repetitions of gradient decent algorithm for the learning. ln this paper, we present a new learning method to overcome the above drawback. The new method realize both fast adaptation and pruning redundant cells by the combination of fast and slaw learning networks.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Incremental Learning / Reduction of Cells / Resource Allocating Network(RAN) / Neural Network
Paper # NC2000-163
Date of Issue

Conference Information
Committee NC
Conference Date 2001/3/16(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) An Incremental Learning and Reduction of Redundant Cells by Sleeping
Sub Title (in English)
Keyword(1) Incremental Learning
Keyword(2) Reduction of Cells
Keyword(3) Resource Allocating Network(RAN)
Keyword(4) Neural Network
1st Author's Name Takahiro TAGAWA
1st Author's Affiliation Department of Intelligence and Computer Science, Nagoya Institute OF Technology()
2nd Author's Name Koichiro YAMAUCHI
2nd Author's Affiliation Information, Electronics and Systems Engineering, Graduate School of Engineering, Hokkaido University.
3rd Author's Name Naohiro ISHII
3rd Author's Affiliation Department of Intelligence and Computer Science, Nagoya Institute OF Technology
4th Author's Name Sachio ITO
4th Author's Affiliation Department of Intelligence and Computer Science, Nagoya Institute OF Technology
Date 2001/3/16
Paper # NC2000-163
Volume (vol) vol.100
Number (no) 688
Page pp.pp.-
#Pages 7
Date of Issue