A high order neural network to solve crossbar switch problem

Yuxin Ding; Li Dong; Ling Wang; Guohua Wu

doi:10.1007/978-3-642-17534-3_85

A high order neural network to solve crossbar switch problem

Yuxin Ding^*
, Li Dong
, Ling Wang
, Guohua Wu

^*Corresponding author for this work

Harbin Institute of Technology Shenzhen
Nanyang Technological University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Scopus citations

Abstract

High-order neural networks can be considered as an expansion of Hopfield neural networks, and have stronger approximation property and faster convergence rate. However, in practice high order network is seldom to be used to solve combinatorial optimization problem. In this paper crossbar switch problem, which is an NP-complete problem, is used as an example to demonstrate how to use high order discrete Hopfield neural network to solve engineering optimization problems. The construction method of energy function and the neural computing algorithm are presented. It is also discussed the method how to speed the convergence and escape from local minima. Experimental results show that high order network has a quick convergence speed, and outperforms the traditional discrete Hopfield network.

Original language	English
Title of host publication	Neural Information Processing
Subtitle of host publication	Models and Applications - 17th International Conference, ICONIP 2010, Proceedings
Pages	692-699
Number of pages	8
Edition	PART 2
DOIs	https://doi.org/10.1007/978-3-642-17534-3_85
State	Published - 2010
Externally published	Yes
Event	17th International Conference on Neural Information Processing, ICONIP 2010 - Sydney, NSW, Australia Duration: 22 Nov 2010 → 25 Nov 2010

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 2
Volume	6444 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	17th International Conference on Neural Information Processing, ICONIP 2010
Country/Territory	Australia
City	Sydney, NSW
Period	22/11/10 → 25/11/10

Keywords

Hopfield network
constraint satisfaction
crossbar switch problem

Access to Document

10.1007/978-3-642-17534-3_85

Cite this

Ding, Y., Dong, L., Wang, L., & Wu, G. (2010). A high order neural network to solve crossbar switch problem. In Neural Information Processing: Models and Applications - 17th International Conference, ICONIP 2010, Proceedings (PART 2 ed., pp. 692-699). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6444 LNCS, No. PART 2). https://doi.org/10.1007/978-3-642-17534-3_85

Ding, Yuxin ; Dong, Li ; Wang, Ling et al. / A high order neural network to solve crossbar switch problem. Neural Information Processing: Models and Applications - 17th International Conference, ICONIP 2010, Proceedings. PART 2. ed. 2010. pp. 692-699 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2).

@inproceedings{d58d0bf2cf754b36952dca7bf750be36,

title = "A high order neural network to solve crossbar switch problem",

abstract = "High-order neural networks can be considered as an expansion of Hopfield neural networks, and have stronger approximation property and faster convergence rate. However, in practice high order network is seldom to be used to solve combinatorial optimization problem. In this paper crossbar switch problem, which is an NP-complete problem, is used as an example to demonstrate how to use high order discrete Hopfield neural network to solve engineering optimization problems. The construction method of energy function and the neural computing algorithm are presented. It is also discussed the method how to speed the convergence and escape from local minima. Experimental results show that high order network has a quick convergence speed, and outperforms the traditional discrete Hopfield network.",

keywords = "Hopfield network, constraint satisfaction, crossbar switch problem",

author = "Yuxin Ding and Li Dong and Ling Wang and Guohua Wu",

year = "2010",

doi = "10.1007/978-3-642-17534-3\_85",

language = "英语",

isbn = "3642175333",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

number = "PART 2",

pages = "692--699",

booktitle = "Neural Information Processing",

edition = "PART 2",

note = "17th International Conference on Neural Information Processing, ICONIP 2010 ; Conference date: 22-11-2010 Through 25-11-2010",

}

Ding, Y, Dong, L, Wang, L & Wu, G 2010, A high order neural network to solve crossbar switch problem. in Neural Information Processing: Models and Applications - 17th International Conference, ICONIP 2010, Proceedings. PART 2 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 2, vol. 6444 LNCS, pp. 692-699, 17th International Conference on Neural Information Processing, ICONIP 2010, Sydney, NSW, Australia, 22/11/10. https://doi.org/10.1007/978-3-642-17534-3_85

A high order neural network to solve crossbar switch problem. / Ding, Yuxin; Dong, Li; Wang, Ling et al.
Neural Information Processing: Models and Applications - 17th International Conference, ICONIP 2010, Proceedings. PART 2. ed. 2010. p. 692-699 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6444 LNCS, No. PART 2).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A high order neural network to solve crossbar switch problem

AU - Ding, Yuxin

AU - Dong, Li

AU - Wang, Ling

AU - Wu, Guohua

PY - 2010

Y1 - 2010

N2 - High-order neural networks can be considered as an expansion of Hopfield neural networks, and have stronger approximation property and faster convergence rate. However, in practice high order network is seldom to be used to solve combinatorial optimization problem. In this paper crossbar switch problem, which is an NP-complete problem, is used as an example to demonstrate how to use high order discrete Hopfield neural network to solve engineering optimization problems. The construction method of energy function and the neural computing algorithm are presented. It is also discussed the method how to speed the convergence and escape from local minima. Experimental results show that high order network has a quick convergence speed, and outperforms the traditional discrete Hopfield network.

AB - High-order neural networks can be considered as an expansion of Hopfield neural networks, and have stronger approximation property and faster convergence rate. However, in practice high order network is seldom to be used to solve combinatorial optimization problem. In this paper crossbar switch problem, which is an NP-complete problem, is used as an example to demonstrate how to use high order discrete Hopfield neural network to solve engineering optimization problems. The construction method of energy function and the neural computing algorithm are presented. It is also discussed the method how to speed the convergence and escape from local minima. Experimental results show that high order network has a quick convergence speed, and outperforms the traditional discrete Hopfield network.

KW - Hopfield network

KW - constraint satisfaction

KW - crossbar switch problem

UR - https://www.scopus.com/pages/publications/78650201021

U2 - 10.1007/978-3-642-17534-3_85

DO - 10.1007/978-3-642-17534-3_85

M3 - 会议稿件

AN - SCOPUS:78650201021

SN - 3642175333

SN - 9783642175336

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 692

EP - 699

BT - Neural Information Processing

T2 - 17th International Conference on Neural Information Processing, ICONIP 2010

Y2 - 22 November 2010 through 25 November 2010

ER -

Ding Y, Dong L, Wang L, Wu G. A high order neural network to solve crossbar switch problem. In Neural Information Processing: Models and Applications - 17th International Conference, ICONIP 2010, Proceedings. PART 2 ed. 2010. p. 692-699. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2). doi: 10.1007/978-3-642-17534-3_85

A high order neural network to solve crossbar switch problem

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this