Policy Gradient with Self-Attention for Model-Free Distributed Nonlinear Multi-Agent Games

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Sebastián, Eduardo, Keskar, Maitrayee, Iqbal, Eeman, Montijano, Eduardo, Sagüés, Carlos, Atanasov, Nikolay
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908552891727872
author Sebastián, Eduardo
Keskar, Maitrayee
Iqbal, Eeman
Montijano, Eduardo
Sagüés, Carlos
Atanasov, Nikolay
author_facet Sebastián, Eduardo
Keskar, Maitrayee
Iqbal, Eeman
Montijano, Eduardo
Sagüés, Carlos
Atanasov, Nikolay
contents Multi-agent games in dynamic nonlinear settings are challenging due to the time-varying interactions among the agents and the non-stationarity of the (potential) Nash equilibria. In this paper we consider model-free games, where agent transitions and costs are observed without knowledge of the transition and cost functions that generate them. We propose a policy gradient approach to learn distributed policies that follow the communication structure in multi-team games, with multiple agents per team. Our formulation is inspired by the structure of distributed policies in linear quadratic games, which take the form of time-varying linear feedback gains. In the nonlinear case, we model the policies as nonlinear feedback gains, parameterized by self-attention layers to account for the time-varying multi-agent communication topology. We demonstrate that our distributed policy gradient approach achieves strong performance in several settings, including distributed linear and nonlinear regulation, and simulated and real multi-robot pursuit-and-evasion games.
format Preprint
id arxiv_https___arxiv_org_abs_2509_18371
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Policy Gradient with Self-Attention for Model-Free Distributed Nonlinear Multi-Agent Games
Sebastián, Eduardo
Keskar, Maitrayee
Iqbal, Eeman
Montijano, Eduardo
Sagüés, Carlos
Atanasov, Nikolay
Systems and Control
Multiagent Systems
Robotics
Multi-agent games in dynamic nonlinear settings are challenging due to the time-varying interactions among the agents and the non-stationarity of the (potential) Nash equilibria. In this paper we consider model-free games, where agent transitions and costs are observed without knowledge of the transition and cost functions that generate them. We propose a policy gradient approach to learn distributed policies that follow the communication structure in multi-team games, with multiple agents per team. Our formulation is inspired by the structure of distributed policies in linear quadratic games, which take the form of time-varying linear feedback gains. In the nonlinear case, we model the policies as nonlinear feedback gains, parameterized by self-attention layers to account for the time-varying multi-agent communication topology. We demonstrate that our distributed policy gradient approach achieves strong performance in several settings, including distributed linear and nonlinear regulation, and simulated and real multi-robot pursuit-and-evasion games.
title Policy Gradient with Self-Attention for Model-Free Distributed Nonlinear Multi-Agent Games
topic Systems and Control
Multiagent Systems
Robotics
url https://arxiv.org/abs/2509.18371