Combining incrementally evolved neural networks based on cellular automata for complex adaptive behaviors

Geum Beom Song; Sung Bae Cho

doi:10.1109/ECNN.2000.886227

Combining incrementally evolved neural networks based on cellular automata for complex adaptive behaviors

Geum Beom Song, Sung Bae Cho

College of Computing

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

3 Citations (Scopus)

Abstract

There has been extensive work to construct an optimal controller for a mobile robot by evolutionary approaches such as genetic algorithm, genetic programming, and so on. However, evolutionary approaches have a difficulty to obtain the controller for complex and general behaviors. In order to overcome this shortcoming, we propose an incremental evolution method for neural networks based on cellular automata (CA) and a method of combining several evolved modules by a rule-based approach. The incremental evolution method evolves the neural network by starting with simpler environment needed simple behavior and gradually making it more complex and general for complex behaviors. The multimodules integration method can make complex and general behaviors by combining several modules evolved or programmed to do simple behavior. Experimental results show the potential of the incremental evolution and multi-modules integration methods as techniques to make the evolved neural network to do complex and general behaviors.

Original language	English
Title of host publication	Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks
Editors	David B. Fogel, Xin Yao
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	121-129
Number of pages	9
ISBN (Electronic)	0780365720, 9780780365728
DOIs	https://doi.org/10.1109/ECNN.2000.886227
Publication status	Published - 2000
Event	1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks, ECNN 2000 - San Antonio, United States Duration: 2000 May 11 → 2000 May 13

Publication series

Name	Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks

Other

Other	1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks, ECNN 2000
Country/Territory	United States
City	San Antonio
Period	00/5/11 → 00/5/13

Bibliographical note

Publisher Copyright:
© 2000 IEEE.

All Science Journal Classification (ASJC) codes

Computer Science(all)

Access to Document

10.1109/ECNN.2000.886227

Cite this

Song, G. B., & Cho, S. B. (2000). Combining incrementally evolved neural networks based on cellular automata for complex adaptive behaviors. In D. B. Fogel, & X. Yao (Eds.), Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks (pp. 121-129). [886227] (Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ECNN.2000.886227

Song, Geum Beom ; Cho, Sung Bae. / Combining incrementally evolved neural networks based on cellular automata for complex adaptive behaviors. Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks. editor / David B. Fogel ; Xin Yao. Institute of Electrical and Electronics Engineers Inc., 2000. pp. 121-129 (Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks).

@inproceedings{fde5a87d81d74c6ba603ae9ee6a7fcaf,

title = "Combining incrementally evolved neural networks based on cellular automata for complex adaptive behaviors",

abstract = "There has been extensive work to construct an optimal controller for a mobile robot by evolutionary approaches such as genetic algorithm, genetic programming, and so on. However, evolutionary approaches have a difficulty to obtain the controller for complex and general behaviors. In order to overcome this shortcoming, we propose an incremental evolution method for neural networks based on cellular automata (CA) and a method of combining several evolved modules by a rule-based approach. The incremental evolution method evolves the neural network by starting with simpler environment needed simple behavior and gradually making it more complex and general for complex behaviors. The multimodules integration method can make complex and general behaviors by combining several modules evolved or programmed to do simple behavior. Experimental results show the potential of the incremental evolution and multi-modules integration methods as techniques to make the evolved neural network to do complex and general behaviors.",

author = "Song, {Geum Beom} and Cho, {Sung Bae}",

note = "Publisher Copyright: {\textcopyright} 2000 IEEE.; 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks, ECNN 2000 ; Conference date: 11-05-2000 Through 13-05-2000",

year = "2000",

doi = "10.1109/ECNN.2000.886227",

language = "English",

series = "Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "121--129",

editor = "Fogel, {David B.} and Xin Yao",

booktitle = "Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks",

address = "United States",

}

Song, GB & Cho, SB 2000, Combining incrementally evolved neural networks based on cellular automata for complex adaptive behaviors. in DB Fogel & X Yao (eds), Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks., 886227, Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks, Institute of Electrical and Electronics Engineers Inc., pp. 121-129, 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks, ECNN 2000, San Antonio, United States, 00/5/11. https://doi.org/10.1109/ECNN.2000.886227

Combining incrementally evolved neural networks based on cellular automata for complex adaptive behaviors. / Song, Geum Beom; Cho, Sung Bae.

Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks. ed. / David B. Fogel; Xin Yao. Institute of Electrical and Electronics Engineers Inc., 2000. p. 121-129 886227 (Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks).

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

TY - GEN

T1 - Combining incrementally evolved neural networks based on cellular automata for complex adaptive behaviors

AU - Song, Geum Beom

AU - Cho, Sung Bae

PY - 2000

Y1 - 2000

N2 - There has been extensive work to construct an optimal controller for a mobile robot by evolutionary approaches such as genetic algorithm, genetic programming, and so on. However, evolutionary approaches have a difficulty to obtain the controller for complex and general behaviors. In order to overcome this shortcoming, we propose an incremental evolution method for neural networks based on cellular automata (CA) and a method of combining several evolved modules by a rule-based approach. The incremental evolution method evolves the neural network by starting with simpler environment needed simple behavior and gradually making it more complex and general for complex behaviors. The multimodules integration method can make complex and general behaviors by combining several modules evolved or programmed to do simple behavior. Experimental results show the potential of the incremental evolution and multi-modules integration methods as techniques to make the evolved neural network to do complex and general behaviors.

AB - There has been extensive work to construct an optimal controller for a mobile robot by evolutionary approaches such as genetic algorithm, genetic programming, and so on. However, evolutionary approaches have a difficulty to obtain the controller for complex and general behaviors. In order to overcome this shortcoming, we propose an incremental evolution method for neural networks based on cellular automata (CA) and a method of combining several evolved modules by a rule-based approach. The incremental evolution method evolves the neural network by starting with simpler environment needed simple behavior and gradually making it more complex and general for complex behaviors. The multimodules integration method can make complex and general behaviors by combining several modules evolved or programmed to do simple behavior. Experimental results show the potential of the incremental evolution and multi-modules integration methods as techniques to make the evolved neural network to do complex and general behaviors.

UR - http://www.scopus.com/inward/record.url?scp=0004158519&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0004158519&partnerID=8YFLogxK

U2 - 10.1109/ECNN.2000.886227

DO - 10.1109/ECNN.2000.886227

M3 - Conference contribution

AN - SCOPUS:0004158519

T3 - Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks

SP - 121

EP - 129

BT - Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks

A2 - Fogel, David B.

A2 - Yao, Xin

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks, ECNN 2000

Y2 - 11 May 2000 through 13 May 2000

ER -

Song GB, Cho SB. Combining incrementally evolved neural networks based on cellular automata for complex adaptive behaviors. In Fogel DB, Yao X, editors, Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks. Institute of Electrical and Electronics Engineers Inc. 2000. p. 121-129. 886227. (Proceedings of the 1st IEEE Symposium on Combinations of Evolutionary Computation and Neural Networks). doi: 10.1109/ECNN.2000.886227

Combining incrementally evolved neural networks based on cellular automata for complex adaptive behaviors

Abstract

Publication series

Other

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this