Presentation 2003/7/24
An Application of Self-Organizing Nural Networks to Module Placement for Gate Array
Noritaka SHIGEl, Hiromi MIYAJIMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this study, applying SOMs to module placement for gate array is discussed. In order to overcome the following difficulties: (i) minimizing wire length, and (ii) locating modules on exact positions, we consider several variations on the following issues: a) probability distribution for input stimuli, b) adaptation rule, and c) re-learning. Our simulation results show that 1) there exists a trade-off between the difficulties (i) and (ii), 2) our re-learning scheme pretty much enhances the difficulty (ii), 3) our probability distribution exhibits better performance than the other distributions, and 4) SOM with re-learning is a moderate solution to the difficulties (i) and (ii).
Keyword(in Japanese) (See Japanese page)
Keyword(in English) self-organizing / neural network / re-larning / module placement / gate array / routing cost
Paper # AI2003-10
Date of Issue

Conference Information
Committee AI
Conference Date 2003/7/24(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 Artificial Intelligence and Knowledge-Based Processing (AI)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) An Application of Self-Organizing Nural Networks to Module Placement for Gate Array
Sub Title (in English)
Keyword(1) self-organizing
Keyword(2) neural network
Keyword(3) re-larning
Keyword(4) module placement
Keyword(5) gate array
Keyword(6) routing cost
1st Author's Name Noritaka SHIGEl
1st Author's Affiliation Faculty of Engineering, Kagoshima University()
2nd Author's Name Hiromi MIYAJIMA
2nd Author's Affiliation Faculty of Engineering, Kagoshima University
Date 2003/7/24
Paper # AI2003-10
Volume (vol) vol.103
Number (no) 243
Page pp.pp.-
#Pages 6
Date of Issue