Permutation free encoding technique for evolving neural networks

Anupam Das; Md Shohrab Hossain; Saeed Muhammad Abdullah; Rashed Ul Islam

doi:10.1007/978-3-540-87732-5_29

Permutation free encoding technique for evolving neural networks

Anupam Das, Md Shohrab Hossain, Saeed Muhammad Abdullah, Rashed Ul Islam

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

2 Scopus citations

Abstract

This paper presents a new evolutionary system using genetic algorithm for evolving artificial neural networks (ANNs). The proposed algorithm is "Permutation free Encoding Technique for Evolving Neural Networks"(PETENN) that uses a novel encoding scheme for representing ANNs. Existing genetic algorithms (GAs) for evolving ANNs suffer from the permutation problem, resulting from the recombination operator. Evolutionary Programming (EP) does not use recombination operator entirely. But the proposed encoding scheme avoids permutation problem by applying a sorting technique. PETENN uses two types of recombination operators that ensure automatic addition or deletion of nodes or links during the crossover process. The evolutionary system has been implemented and applied to a number of benchmark problems in machine learning and neural networks. The experimental results show that the system can dynamically evolve ANN architectures, showing competitiveness and, in some cases, superiority in performance.

Original language	English (US)
Title of host publication	Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings
Publisher	Springer Verlag
Pages	255-265
Number of pages	11
Edition	PART 1
ISBN (Print)	3540877312, 9783540877318
DOIs	https://doi.org/10.1007/978-3-540-87732-5_29
State	Published - 2008
Event	5th International Symposium on Neural Networks, ISNN 2008 - Beijing, China Duration: Sep 24 2008 → Sep 28 2008

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 1
Volume	5263 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	5th International Symposium on Neural Networks, ISNN 2008
Country/Territory	China
City	Beijing
Period	9/24/08 → 9/28/08

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
General Computer Science

Access to Document

10.1007/978-3-540-87732-5_29

Cite this

Das, A., Hossain, M. S., Abdullah, S. M., & Ul Islam, R. (2008). Permutation free encoding technique for evolving neural networks. In Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings (PART 1 ed., pp. 255-265). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5263 LNCS, No. PART 1). Springer Verlag. https://doi.org/10.1007/978-3-540-87732-5_29

Das, Anupam ; Hossain, Md Shohrab ; Abdullah, Saeed Muhammad et al. / Permutation free encoding technique for evolving neural networks. Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings. PART 1. ed. Springer Verlag, 2008. pp. 255-265 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1).

@inproceedings{7f310ba3d2094a89bedbae9248c85784,

title = "Permutation free encoding technique for evolving neural networks",

abstract = "This paper presents a new evolutionary system using genetic algorithm for evolving artificial neural networks (ANNs). The proposed algorithm is {"}Permutation free Encoding Technique for Evolving Neural Networks{"}(PETENN) that uses a novel encoding scheme for representing ANNs. Existing genetic algorithms (GAs) for evolving ANNs suffer from the permutation problem, resulting from the recombination operator. Evolutionary Programming (EP) does not use recombination operator entirely. But the proposed encoding scheme avoids permutation problem by applying a sorting technique. PETENN uses two types of recombination operators that ensure automatic addition or deletion of nodes or links during the crossover process. The evolutionary system has been implemented and applied to a number of benchmark problems in machine learning and neural networks. The experimental results show that the system can dynamically evolve ANN architectures, showing competitiveness and, in some cases, superiority in performance.",

author = "Anupam Das and Hossain, {Md Shohrab} and Abdullah, {Saeed Muhammad} and {Ul Islam}, Rashed",

year = "2008",

doi = "10.1007/978-3-540-87732-5_29",

language = "English (US)",

isbn = "3540877312",

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

publisher = "Springer Verlag",

number = "PART 1",

pages = "255--265",

booktitle = "Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings",

address = "Germany",

edition = "PART 1",

}

Das, A, Hossain, MS, Abdullah, SM & Ul Islam, R 2008, Permutation free encoding technique for evolving neural networks. in Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings. PART 1 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 1, vol. 5263 LNCS, Springer Verlag, pp. 255-265, 5th International Symposium on Neural Networks, ISNN 2008, Beijing, China, 9/24/08. https://doi.org/10.1007/978-3-540-87732-5_29

Permutation free encoding technique for evolving neural networks. / Das, Anupam; Hossain, Md Shohrab; Abdullah, Saeed Muhammad et al.
Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings. PART 1. ed. Springer Verlag, 2008. p. 255-265 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5263 LNCS, No. PART 1).

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

TY - GEN

T1 - Permutation free encoding technique for evolving neural networks

AU - Das, Anupam

AU - Hossain, Md Shohrab

AU - Abdullah, Saeed Muhammad

AU - Ul Islam, Rashed

PY - 2008

Y1 - 2008

N2 - This paper presents a new evolutionary system using genetic algorithm for evolving artificial neural networks (ANNs). The proposed algorithm is "Permutation free Encoding Technique for Evolving Neural Networks"(PETENN) that uses a novel encoding scheme for representing ANNs. Existing genetic algorithms (GAs) for evolving ANNs suffer from the permutation problem, resulting from the recombination operator. Evolutionary Programming (EP) does not use recombination operator entirely. But the proposed encoding scheme avoids permutation problem by applying a sorting technique. PETENN uses two types of recombination operators that ensure automatic addition or deletion of nodes or links during the crossover process. The evolutionary system has been implemented and applied to a number of benchmark problems in machine learning and neural networks. The experimental results show that the system can dynamically evolve ANN architectures, showing competitiveness and, in some cases, superiority in performance.

AB - This paper presents a new evolutionary system using genetic algorithm for evolving artificial neural networks (ANNs). The proposed algorithm is "Permutation free Encoding Technique for Evolving Neural Networks"(PETENN) that uses a novel encoding scheme for representing ANNs. Existing genetic algorithms (GAs) for evolving ANNs suffer from the permutation problem, resulting from the recombination operator. Evolutionary Programming (EP) does not use recombination operator entirely. But the proposed encoding scheme avoids permutation problem by applying a sorting technique. PETENN uses two types of recombination operators that ensure automatic addition or deletion of nodes or links during the crossover process. The evolutionary system has been implemented and applied to a number of benchmark problems in machine learning and neural networks. The experimental results show that the system can dynamically evolve ANN architectures, showing competitiveness and, in some cases, superiority in performance.

UR - http://www.scopus.com/inward/record.url?scp=59149085211&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=59149085211&partnerID=8YFLogxK

U2 - 10.1007/978-3-540-87732-5_29

DO - 10.1007/978-3-540-87732-5_29

M3 - Conference contribution

AN - SCOPUS:59149085211

SN - 3540877312

SN - 9783540877318

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

SP - 255

EP - 265

BT - Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings

PB - Springer Verlag

T2 - 5th International Symposium on Neural Networks, ISNN 2008

Y2 - 24 September 2008 through 28 September 2008

ER -

Das A, Hossain MS, Abdullah SM, Ul Islam R. Permutation free encoding technique for evolving neural networks. In Advances in Neural Networks - ISNN 2008 - 5th International Symposium on Neural Networks, ISNN 2008, Proceedings. PART 1 ed. Springer Verlag. 2008. p. 255-265. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 1). doi: 10.1007/978-3-540-87732-5_29

Permutation free encoding technique for evolving neural networks

Abstract

Publication series

Other

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this