Presentation 1994/1/21
Applications of Neural Network Which Includes Sigma Pai Units to Optimization
Sadayuki Murashima, Takayuki Toyohira, Takayasu Fuchida, Hiromi Miyajima,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) A formulation of neural network for solving traveling salesman problems is discussed.The formulation is based on the city adjacency in the tour and has an advantage of rapid convergence to the optimal state.However it is well known that the networks often converge to an invalid solution which contains closed subtours.In order to suppress the closed subtours by using higher order terms, we discuss the applicability of neural network which has an energy function of higher order to optimization.Numerical simulations show that the generation of closed subtours are suppressed effectively in the network including higher order terms.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Hopfield Network / Energy Function / Higher Order Term / Optimization
Paper # CAS93-103
Date of Issue

Conference Information
Committee CAS
Conference Date 1994/1/21(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 Circuits and Systems (CAS)
Language ENG
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Applications of Neural Network Which Includes Sigma Pai Units to Optimization
Sub Title (in English)
Keyword(1) Hopfield Network
Keyword(2) Energy Function
Keyword(3) Higher Order Term
Keyword(4) Optimization
1st Author's Name Sadayuki Murashima
1st Author's Affiliation Faculty of Engineering,Kagoshima University()
2nd Author's Name Takayuki Toyohira
2nd Author's Affiliation Kagoshima Techinical College
3rd Author's Name Takayasu Fuchida
3rd Author's Affiliation Faculty of Engineering,Kagoshima University
4th Author's Name Hiromi Miyajima
4th Author's Affiliation Faculty of Engineering,Information and Computer Science,Kagoshima University
Date 1994/1/21
Paper # CAS93-103
Volume (vol) vol.93
Number (no) 432
Page pp.pp.-
#Pages 7
Date of Issue