Darwinian dynamics of embodied chaotic exploration

Shim, Yoonsik, Auerbach, Joshua E and Husbands, Phil (2016) Darwinian dynamics of embodied chaotic exploration. GECCO'16: Genetic and Evolutionary Computation Conference: A Recombination of the 25th Conference on Genetic Algorithms (IGCA) and the 21st Annual Genetic Programming Conference (GP), Denver, Colorado, July 20-24 2016. Published in: Friedrich, Tobias, (ed.) Proceedings of the 2016 on Genetic and Evolutionary Computation Conference Companion. 1053-1056. Association for Computing Machinery ISBN 9781450343237

Full text not available from this repository.

Abstract

We introduce a complementary direction of research into
another kind of possible candidate for Darwinian neural dynamics where such dynamics occur at a slightly higher, more
abstract level. Crucially the (chaotic) neuro dynamics are
embodied and it is the whole neuro-body-environment system
that must be considered, although the changes occur
at the neural level. The Embodied Chaotic Exploration
(ECE) incrementally explores and learns motor behaviors
through an integrated combination of chaotic search and
reflex learning. The architecture developed here allows realtime
goal-directed exploration and learning of the possible
motor patterns (e.g., for locomotion) of embodied systems
of arbitrary morphology. The overall iterative search process
formed from this combination is shown to have a strong
parallels with evolutionary search.

Item Type: Conference Proceedings
Schools and Departments: School of Engineering and Informatics > Informatics
Research Centres and Groups: Centre for Computational Neuroscience and Robotics
Subjects: Q Science > QA Mathematics > QA0075 Electronic computers. Computer science
Depositing User: Phil Husbands
Date Deposited: 17 Oct 2018 11:52
Last Modified: 17 Oct 2018 11:52
URI: http://srodev.sussex.ac.uk/id/eprint/79547
📧 Request an update