Open Ad Hoc Teamwork with Cooperative Game Theory

Loading...
Thumbnail Image

Access rights

openAccess
CC BY
publishedVersion

URL

Journal Title

Journal ISSN

Volume Title

A4 Artikkeli konferenssijulkaisussa

Date

Major/Subject

Mcode

Degree programme

Language

en

Pages

29

Series

Proceedings of the 41st International Conference on Machine Learning, Volume 235, pp. 50902-50930, Proceedings of Machine Learning Research ; Volume 235

Abstract

Ad hoc teamwork poses a challenging problem, requiring the design of an agent to collaborate with teammates without prior coordination or joint training. Open ad hoc teamwork (OAHT) further complicates this challenge by considering environments with a changing number of teammates, referred to as open teams. One promising solution in practice to this problem is leveraging the generalizability of graph neural networks to handle an unrestricted number of agents with various agent-types, named graph-based policy learning (GPL). However, its joint Q-value representation over a coordination graph lacks convincing explanations. In this paper, we establish a new theory to understand the representation of the joint Q-value for OAHT and its learning paradigm, through the lens of cooperative game theory. Building on our theory, we propose a novel algorithm named CIAO, based on GPL’s framework, with additional provable implementation tricks that can facilitate learning. The demos of experimental results are available on https://sites.google.com/view/ciao2024, and the code of experiments is published on https://github.com/hsvgbkhgbv/CIAO.

Description

Keywords

Other note

Citation

Wang, J, Li, Y, Zhang, Y, Pan, W & Kaski, S 2024, Open Ad Hoc Teamwork with Cooperative Game Theory. in R Salakhutdinov, Z Kolter, K Heller, A Weller, N Oliver, J Scarlett & F Berkenkamp (eds), Proceedings of the 41st International Conference on Machine Learning. vol. 235, Proceedings of Machine Learning Research, vol. 235, JMLR, pp. 50902-50930, International Conference on Machine Learning, Vienna, Austria, 21/07/2024. < https://arxiv.org/abs/2402.15259 >