Does Training Lead to the Formation of Modules in Threshold Networks?

Research output: Contribution in Book/Catalog/Report/Conference proceedingConference contribution

18 Downloads (Pure)

Abstract

This paper addresses the question to determine the necessary
conditions for the emergence of modules in the framework of artificial
evolution. In particular, threshold networks are trained as controllers
for robots able to perform two different tasks at the same time. It is
shown that modules do not emerge under a wide set of conditions in our
experimental framework. This finding supports the hypothesis that the
emergence of modularity indeed depends upon the algorithm used for
artificial evolution and the characteristics of the tasks.
Original languageEnglish
Title of host publicationProceedings of ECCS 2014
Subtitle of host publicationEuropean Conference on Complex Systems
PublisherSpringer
Pages181-192
Number of pages12
ISBN (Electronic)978-3-319-29228-1
ISBN (Print)978-3-319-29226-7
Publication statusPublished - 1 May 2016
Eventeccs'14 - Lucca, Italy
Duration: 22 Sep 201426 Sep 2014

Publication series

NameSpringer Proceedings in Complexity
PublisherSpringer

Scientific committee

Scientific committeeeccs'14
CountryItaly
CityLucca
Period22/09/1426/09/14

Fingerprint

Robots

Keywords

  • neural networks
  • learning
  • genetic algorithm
  • modularity
  • evolutionary robotics

Cite this

Nicolay, D., Andrea, R., & Carletti, T. (2016). Does Training Lead to the Formation of Modules in Threshold Networks? In Proceedings of ECCS 2014: European Conference on Complex Systems (pp. 181-192). (Springer Proceedings in Complexity). Springer.
Nicolay, Delphine ; Andrea, Roli ; Carletti, Timoteo. / Does Training Lead to the Formation of Modules in Threshold Networks?. Proceedings of ECCS 2014: European Conference on Complex Systems. Springer, 2016. pp. 181-192 (Springer Proceedings in Complexity).
@inproceedings{8e1ebee40c7647a0a129e5e95d1c96e4,
title = "Does Training Lead to the Formation of Modules in Threshold Networks?",
abstract = "This paper addresses the question to determine the necessaryconditions for the emergence of modules in the framework of artificialevolution. In particular, threshold networks are trained as controllersfor robots able to perform two different tasks at the same time. It isshown that modules do not emerge under a wide set of conditions in ourexperimental framework. This finding supports the hypothesis that theemergence of modularity indeed depends upon the algorithm used forartificial evolution and the characteristics of the tasks.",
keywords = "neural networks, learning, genetic algorithm, modularity, evolutionary robotics",
author = "Delphine Nicolay and Roli Andrea and Timoteo Carletti",
year = "2016",
month = "5",
day = "1",
language = "English",
isbn = "978-3-319-29226-7",
series = "Springer Proceedings in Complexity",
publisher = "Springer",
pages = "181--192",
booktitle = "Proceedings of ECCS 2014",

}

Nicolay, D, Andrea, R & Carletti, T 2016, Does Training Lead to the Formation of Modules in Threshold Networks? in Proceedings of ECCS 2014: European Conference on Complex Systems. Springer Proceedings in Complexity, Springer, pp. 181-192, eccs'14, Lucca, Italy, 22/09/14.

Does Training Lead to the Formation of Modules in Threshold Networks? / Nicolay, Delphine; Andrea, Roli; Carletti, Timoteo.

Proceedings of ECCS 2014: European Conference on Complex Systems. Springer, 2016. p. 181-192 (Springer Proceedings in Complexity).

Research output: Contribution in Book/Catalog/Report/Conference proceedingConference contribution

TY - GEN

T1 - Does Training Lead to the Formation of Modules in Threshold Networks?

AU - Nicolay, Delphine

AU - Andrea, Roli

AU - Carletti, Timoteo

PY - 2016/5/1

Y1 - 2016/5/1

N2 - This paper addresses the question to determine the necessaryconditions for the emergence of modules in the framework of artificialevolution. In particular, threshold networks are trained as controllersfor robots able to perform two different tasks at the same time. It isshown that modules do not emerge under a wide set of conditions in ourexperimental framework. This finding supports the hypothesis that theemergence of modularity indeed depends upon the algorithm used forartificial evolution and the characteristics of the tasks.

AB - This paper addresses the question to determine the necessaryconditions for the emergence of modules in the framework of artificialevolution. In particular, threshold networks are trained as controllersfor robots able to perform two different tasks at the same time. It isshown that modules do not emerge under a wide set of conditions in ourexperimental framework. This finding supports the hypothesis that theemergence of modularity indeed depends upon the algorithm used forartificial evolution and the characteristics of the tasks.

KW - neural networks

KW - learning

KW - genetic algorithm

KW - modularity

KW - evolutionary robotics

UR - http://www.springer.com/gp/book/9783319292267

M3 - Conference contribution

SN - 978-3-319-29226-7

T3 - Springer Proceedings in Complexity

SP - 181

EP - 192

BT - Proceedings of ECCS 2014

PB - Springer

ER -

Nicolay D, Andrea R, Carletti T. Does Training Lead to the Formation of Modules in Threshold Networks? In Proceedings of ECCS 2014: European Conference on Complex Systems. Springer. 2016. p. 181-192. (Springer Proceedings in Complexity).