English

An active learning method for solving competitive multi-agent decision-making and control problems

Systems and Control 2024-10-10 v5 Machine Learning Multiagent Systems Systems and Control Optimization and Control

Abstract

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.

Keywords

Cite

@article{arxiv.2212.12561,
  title  = {An active learning method for solving competitive multi-agent decision-making and control problems},
  author = {Filippo Fabiani and Alberto Bemporad},
  journal= {arXiv preprint arXiv:2212.12561},
  year   = {2024}
}

Comments

Python package available at https://github.com/bemporad/gnep-learn

R2 v1 2026-06-28T07:51:15.613Z