Building collaboration in multi-agent systems using reinforcement learning

Aydin, Mehmet Emin; Fellows, Ryan

Abstract

© Springer Nature Switzerland AG 2018. This paper presents a proof-of concept study for demonstrating the viability of building collaboration among multiple agents through standard Q learning algorithm embedded in particle swarm optimisation. Collaboration is formulated to be achieved among the agents via competition, where the agents are expected to balance their action in such a way that none of them drifts away of the team and none intervene any fellow neighbours territory, either. Particles are devised with Q learning for self training to learn how to act as members of a swarm and how to produce collaborative/collective behaviours. The produced experimental results are supportive to the proposed idea suggesting that a substantive collaboration can be build via proposed learning algorithm.

Citation

Aydin, M. E., & Fellows, R. (2018). Building collaboration in multi-agent systems using reinforcement learning. Lecture Notes in Artificial Intelligence, 11056 LNAI, 201-212. https://doi.org/10.1007/978-3-319-98446-9_19

• Journal Article Type: Conference Paper
• Conference Name: 10th International Conference, ICCCI 2018
• Start Date: Sep 5, 2018
• End Date: Sep 7, 2018
• Acceptance Date: Jun 10, 2018
• Online Publication Date: Aug 8, 2018
• Publication Date: Jan 1, 2018
• Journal: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
• Print ISSN: 0302-9743
• Electronic ISSN: 1611-3349
• Publisher: Springer Verlag
• Peer Reviewed: Peer Reviewed
• Volume: 11056 LNAI
• Pages: 201-212
• Book Title: Computational Collective Intelligence
• ISBN: 9783319984452
• DOI: https://doi.org/10.1007/978-3-319-98446-9_19
• Keywords: agent collaboration, reinforcement learning, multi-agent systems, Q learning, disaster management
• Public URL: https://uwe-repository.worktribe.com/output/862496
• Publisher URL: https://doi.org/10.1007/978-3-319-98446-9_19
• Additional Information: Additional Information : The final authenticated version is available online at https://doi.org/10.1007/978-3-319-98446-9_19; Title of Conference or Conference Proceedings : 10th International Conference, ICCCI 2018, Bristol, UK