Evolutionary algorithms for real-time artificial neural network training

Ananda Jagadeesan; Grant Maxwell; Christopher MacLeod

Evolutionary algorithms for real-time artificial neural network training

Ananda Jagadeesan, Grant Maxwell, Christopher MacLeod

UHI North West and Hebrides

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

1 Citation (Scopus)

Abstract

This paper reports on experiments investigating the use of Evolutionary Algorithms to train Artificial Neural Networks in real time. A simulated legged mobile robot was used as a test bed in the experiments. Since the algorithm is designed to be used with a physical robot, the population size was one and the recombination operator was not used. The algorithm is therefore rather similar to the original Evolutionary Strategies concept. The idea is that such an algorithm could eventually be used to alter the locomotive performance of the robot on different terrain types. Results are presented showing the effect of various algorithm parameters on system performance.

Original language	English
Title of host publication	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages	73-78
Number of pages	6
Publication status	Published - 2005
Event	15th International Conference on Artificial Neural Networks: Biological Inspirations - ICANN 2005 - Warsaw, Poland Duration: 11 Sept 2005 → 15 Sept 2005

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	3697 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	15th International Conference on Artificial Neural Networks: Biological Inspirations - ICANN 2005
Country/Territory	Poland
City	Warsaw
Period	11/09/05 → 15/09/05

Cite this

Jagadeesan, A., Maxwell, G., & MacLeod, C. (2005). Evolutionary algorithms for real-time artificial neural network training. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 73-78). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3697 LNCS).

Jagadeesan, Ananda ; Maxwell, Grant ; MacLeod, Christopher. / Evolutionary algorithms for real-time artificial neural network training. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 2005. pp. 73-78 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{6e7347f639324ed694ccac707cf37cae,

title = "Evolutionary algorithms for real-time artificial neural network training",

abstract = "This paper reports on experiments investigating the use of Evolutionary Algorithms to train Artificial Neural Networks in real time. A simulated legged mobile robot was used as a test bed in the experiments. Since the algorithm is designed to be used with a physical robot, the population size was one and the recombination operator was not used. The algorithm is therefore rather similar to the original Evolutionary Strategies concept. The idea is that such an algorithm could eventually be used to alter the locomotive performance of the robot on different terrain types. Results are presented showing the effect of various algorithm parameters on system performance.",

author = "Ananda Jagadeesan and Grant Maxwell and Christopher MacLeod",

year = "2005",

language = "English",

isbn = "3540287558",

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

pages = "73--78",

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

note = "15th International Conference on Artificial Neural Networks: Biological Inspirations - ICANN 2005 ; Conference date: 11-09-2005 Through 15-09-2005",

}

Jagadeesan, A, Maxwell, G & MacLeod, C 2005, Evolutionary algorithms for real-time artificial neural network training. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 3697 LNCS, pp. 73-78, 15th International Conference on Artificial Neural Networks: Biological Inspirations - ICANN 2005, Warsaw, Poland, 11/09/05.

Evolutionary algorithms for real-time artificial neural network training. / Jagadeesan, Ananda; Maxwell, Grant; MacLeod, Christopher.
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 2005. p. 73-78 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3697 LNCS).

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

TY - GEN

T1 - Evolutionary algorithms for real-time artificial neural network training

AU - Jagadeesan, Ananda

AU - Maxwell, Grant

AU - MacLeod, Christopher

PY - 2005

Y1 - 2005

N2 - This paper reports on experiments investigating the use of Evolutionary Algorithms to train Artificial Neural Networks in real time. A simulated legged mobile robot was used as a test bed in the experiments. Since the algorithm is designed to be used with a physical robot, the population size was one and the recombination operator was not used. The algorithm is therefore rather similar to the original Evolutionary Strategies concept. The idea is that such an algorithm could eventually be used to alter the locomotive performance of the robot on different terrain types. Results are presented showing the effect of various algorithm parameters on system performance.

AB - This paper reports on experiments investigating the use of Evolutionary Algorithms to train Artificial Neural Networks in real time. A simulated legged mobile robot was used as a test bed in the experiments. Since the algorithm is designed to be used with a physical robot, the population size was one and the recombination operator was not used. The algorithm is therefore rather similar to the original Evolutionary Strategies concept. The idea is that such an algorithm could eventually be used to alter the locomotive performance of the robot on different terrain types. Results are presented showing the effect of various algorithm parameters on system performance.

UR - http://www.scopus.com/inward/record.url?scp=33646242770&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33646242770&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:33646242770

SN - 3540287558

SN - 9783540287551

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

SP - 73

EP - 78

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

T2 - 15th International Conference on Artificial Neural Networks: Biological Inspirations - ICANN 2005

Y2 - 11 September 2005 through 15 September 2005

ER -

Jagadeesan A, Maxwell G, MacLeod C. Evolutionary algorithms for real-time artificial neural network training. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). 2005. p. 73-78. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

Evolutionary algorithms for real-time artificial neural network training

Abstract

Publication series

Conference

Other files and links

Fingerprint

Cite this