An active learning method for solving competitive multi-agent decision-making and control problems
This addresses decision-making and control challenges in competitive multi-agent environments, but it appears incremental as it builds on existing active learning and estimation methods.
The paper tackles the problem of finding stationary action profiles in competitive multi-agent systems by introducing an active-learning scheme where a central observer probes agents and updates local estimates, establishing conditions for convergence and existence of such profiles, with numerical simulations demonstrating effectiveness.
To identify a stationary action profile for a population of competitive agents, each executing private strategies, we introduce a novel active-learning scheme where a centralized external observer (or entity) can probe the agents' reactions and recursively update simple local parametric estimates of the action-reaction mappings. Under very general working assumptions (not even assuming that a stationary profile exists), sufficient conditions are established to assess the asymptotic properties of the proposed active learning methodology so that, if the parameters characterizing the action-reaction mappings converge, a stationary action profile is achieved. Such conditions hence act also as certificates for the existence of such a profile. Extensive numerical simulations involving typical competitive multi-agent control and decision-making problems illustrate the practical effectiveness of the proposed learning-based approach.