Skip to Main content Skip to Navigation
Journal articles

Neural reinforcement learning for behaviour synthesis

Abstract : We present the results of a research aimed at improving the Q-learning method through the use of artificial neural networks. Neural implementations are interesting due to their generalisation ability. Two implementations are proposed: one with a competitive multilayer perceptron and the other with a self-organising map. Results obtained on a task of learning an obstacle avoidance behaviour for the mobile miniature robot Khepera show that this last implementation is very effective, learning more than 40 times faster than the basic Q-learning implementation. These neural implementations are also compared with several Q-learning enhancements, like the Q-learning with Hamming distance, Q-learning with statistical clustering and Dyna-Q.
Complete list of metadatas

Cited literature [26 references]  Display  Hide  Download

https://hal-amu.archives-ouvertes.fr/hal-01337989
Contributor : Claude Touzet <>
Submitted on : Monday, June 27, 2016 - 4:30:28 PM
Last modification on : Monday, January 29, 2018 - 4:48:05 PM

File

Jars_97(1).pdf
Files produced by the author(s)

Identifiers

Collections

Citation

Claude Touzet. Neural reinforcement learning for behaviour synthesis. Robotics and Autonomous Systems, Elsevier, 1997, ⟨10.1016/S0921-8890(97)00042-0⟩. ⟨hal-01337989⟩

Share

Metrics

Record views

207

Files downloads

829