Escaping local optima in multi-agent oriented constraint satisfaction.
Basharu, Muhammed; Ahriz, Hatem; Arana, Inés
We present a multi-agent approach to constraint satisfaction where feedback and reinforcement are used in order to avoid local optima and, consequently, to improve the overall solution. Our approach, FeReRA, is based on the fact that an agents local best performance does not necessarily contribute to the systems best performance. Thus, agents may be rewarded for improving the systems performance and penalised for not contributing towards a better solution. Hence, agents may be forced to choose sub-optimal moves when they reach a specified penalty threshold as a consequence of their lack of contribution towards a better overall solution. This may allow other agents to choose better moves and, therefore, to improve the overall performance of the system. FeReRA is tested against its predecessor, ERA, and a comparative evaluation of both approaches is presented.
|Start Date||Dec 15, 2003|
|Publication Date||Dec 31, 2004|
|Publisher||Springer (part of Springer Nature)|
|Institution Citation||BASHARU, M., AHRIZ, H. and ARANA, I. 2004. Escaping local optima in multi-agent oriented constraint satisfaction. In Coenen, F., Preece, A. and Macintosh, A. (eds.) Research and development in intelligent systems XX: technical proceedings of the 23rd Annual international conference of the British Computer Society's Specialist Group on Artificial Intelligence (SGAI) (AI-2003), 15-17 December 2003, Cambridge, UK. London: Springer [online], pages 97-110. Available from: https://doi.org/10.1007/978-0-85729-412-8_8|
BASHARU 2003 Escaping local optima
You might also like
Dynamic agent prioritisation with penalties in distributed local search.
Plan recommendation for well engineering.
A hybrid approach to solving coarse-grained DisCSPs.
Presentation / Conference