Coordination in Ad Hoc Teams using Knowledge-based Reasoning and Learning

Ad hoc teamwork focuses on enabling an agent to collaborate with others without prior coordination. This is a central problem in the coordination of open multiagent systems in many practical applications. Data-driven learning methods represent the state of the art in ad hoc teamwork. They use a large labeled dataset of prior observations to model the behavior of other agent types and to determine the ad hoc agent’s behavior. These methods are computationally expensive, lack transparency, and make it difficult to adapt to changes in team composition. Our recent work introduced an architecture that determined an ad hoc agent’s behavior based on non-monotonic logical reasoning with prior commonsense domain knowledge and models that were learned from limited examples to predict other agents’ behavior. In this paper, we expand the architecture’s capabilities, supporting: (a) online adaptation and choice of learned models of other agents’ behavior; and (b) collaboration in the presence of partial observability and limited communication. Experimental evaluation in two simulated benchmark domains for ad hoc teamwork demonstrates performance comparable or better than state of the art data-driven baselines in simple and complex scenarios, particularly in the presence of limited training data, partial observability, and changes in team composition.