Towards Open Ad Hoc Teamwork Using Graph-based Policy Learning
This addresses the challenge of flexible collaboration in multi-agent systems for applications like robotics or gaming, but it is incremental as it builds on prior work in ad hoc teamwork by extending it to open teams.
The paper tackles the problem of open ad hoc teamwork, where agents must adapt to teammates entering and leaving without prior coordination, by proposing a graph-based policy learning method that integrates agent and joint-action value models. The result shows that this approach robustly adapts to dynamic team compositions and significantly outperforms alternative methods, though no concrete numbers are provided.
Ad hoc teamwork is the challenging problem of designing an autonomous agent which can adapt quickly to collaborate with teammates without prior coordination mechanisms, including joint training. Prior work in this area has focused on closed teams in which the number of agents is fixed. In this work, we consider open teams by allowing agents with different fixed policies to enter and leave the environment without prior notification. Our solution builds on graph neural networks to learn agent models and joint-action value models under varying team compositions. We contribute a novel action-value computation that integrates the agent model and joint-action value model to produce action-value estimates. We empirically demonstrate that our approach successfully models the effects other agents have on the learner, leading to policies that robustly adapt to dynamic team compositions and significantly outperform several alternative methods.