Search space restriction of neuro-evolution through constrained modularization of neural networks

Autor(en): Rempis, C.W.
Pasemann, F.
Stichwörter: Behavior control; Constraint functions; Crossover operator; Domain knowledge; Input and outputs; Local processing; Modularizations; Non destructive; Non-trivial; Novel techniques; Search spaces; Sensors and actuators; Spatial relations; Synaptic connections, Actuators; Behavioral research; Data processing; Modular construction; Sensors, Recurrent neural networks
Erscheinungsdatum: 2010
Journal: Proceedings of the 6th International Workshop on Artificial Neural Networks and Intelligent Information Processing - Workshop, ANNIIP 2010, in Conjunction with ICINCO 2010
Startseite: 13
Seitenende: 22
Evolving recurrent neural networks for behavior control of robots equipped with larger sets of sensors and actuators is difficult due to the large search spaces that come with the larger number of input and output neurons. We propose constrained modularization as a novel technique to reduce the search space for such evolutions. Appropriate neural networks are divided manually into logically and spatially related neuro-modules based on domain knowledge of the targeted problem. Then constraint functions are applied to these neuro-modules to force the compliance of user defined restrictions and relations. For neuro-modules this will facilitate complex symmetries and other spatial relations, local processing of related sensors and actuators, the reuse of functional neuro-modules, fine control of synaptic connections, and a non-destructive crossover operator. With an implementation of this so called ICONE method several behaviors for non-trivial robots have already been evolved successfully.
Conference of 6th International Workshop on Artificial Neural Networks and Intelligent Information Processing, ANNIIP 2010, in Conjunction with ICINCO 2010 ; Conference Date: 17 June 2010 Through 18 June 2010; Conference Code:82685
ISBN: 9789898425034
Externe URL:

Show full item record

Page view(s)

Last Week
Last month
checked on Mar 5, 2024

Google ScholarTM