Title
Flexible and multistable pattern generation by evolving constrained plastic neurocontrollers
Abstract
In evolutionary robotics, plastic neural network models proved to be promising for evolving adaptive behaviors. In particular, neurocontrollers incorporating hebbian synapses have been shown to be useful for implementing conflicting sub-behaviors. Numerous interesting complex tasks assume such flexibility. However, those evolved controllers often exhibit behavioral instability, as simulation time is extended beyond the short limit used during evolution. In this paper, we propose constrained plastic models inspired by neural homeostasis phenomena, in order to evolve flexible and stable pattern generators for single-legged locomotion. Comparative results show that constrained controllers perform better than unconstrained ones in both terms of evolvability and behavioral stability. Functional analyses of the best evolved controller unveil the adaptivity, robustness and homeostasis arising from the statically constrained plasticity. Interestingly, homeostasis evolved implicitly without relying on any active homeostatic mechanisms and is implemented through hebbian plasticity, usually considered destabilizing.
Year
DOI
Venue
2011
10.1177/1059712311403631
Adaptive Behaviour
Keywords
Field
DocType
adaptive behavior,neural homeostasis phenomenon,plastic model,behavioral stability,comparative result,hebbian plasticity,plastic neural network model,behavioral instability,multistable pattern generation,active homeostatic mechanism,plastic neurocontrollers,hebbian synapsis,homeostasis,neural network,neural network model,evolutionary robotics
Control theory,Evolutionary robotics,Evolvability,Computer science,Robustness (computer science),Hebbian theory,Artificial intelligence,Artificial neural network,Adaptive behavior,Machine learning,Plasticity
Journal
Volume
Issue
ISSN
19
3
1059-7123
Citations 
PageRank 
References 
7
0.46
16
Authors
3
Name
Order
Citations
PageRank
Thierry Hoinville1746.50
Cecilia Tapia Siles270.46
Patrick Henaff37711.33