DocumentCode :
2579319
Title :
Evolvability of Neuromodulated Learning for Robots
Author :
Durr, Peter ; Mattiussi, Claudio ; Soltoggio, Andrea ; Floreano, Dario
Author_Institution :
Lab. of Intell. Syst., Ecole Polytech. Fed. de Lausanne, Lausanne
fYear :
2008
fDate :
6-8 Aug. 2008
Firstpage :
41
Lastpage :
46
Abstract :
Neuromodulation is thought to be one of the underlying principles of learning and memory in biological neural networks. Recent experiments have shown that neuroevolutionary methods benefit from neuromodulation in simple grid-world problems. In this paper we investigate the performance of a neuroevolutionary method applied to a more realistic robotic task. While confirming the favorable effect of neuromodulatory structures, our results indicate that the evolution of such architectures requires a mechanism which allows for selective modular targetting of the neuromodulatory connections.
Keywords :
learning (artificial intelligence); neural nets; robots; biological neural networks; neuroevolutionary methods; neuromodulated learning; robots; Biological systems; Infrared sensors; Intelligent networks; Intelligent robots; Intelligent systems; Laboratories; Neural networks; Neurons; Robot sensing systems; Turning; Learning; Neural Networks; Neuroevolution; Neuromodulation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Learning and Adaptive Behaviors for Robotic Systems, 2008. LAB-RS '08. ECSIS Symposium on
Conference_Location :
Edinburgh
Print_ISBN :
978-0-7695-3272-1
Type :
conf
DOI :
10.1109/LAB-RS.2008.22
Filename :
4599425
Link To Document :
بازگشت