Learning Equilibria from Data: Provably Efficient Multi-Agent Imitation Learning

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Freihaut, Till, Viano, Luca, Cevher, Volkan, Geist, Matthieu, Ramponi, Giorgia
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914083200040960
author Freihaut, Till
Viano, Luca
Cevher, Volkan
Geist, Matthieu
Ramponi, Giorgia
author_facet Freihaut, Till
Viano, Luca
Cevher, Volkan
Geist, Matthieu
Ramponi, Giorgia
contents This paper provides the first expert sample complexity characterization for learning a Nash equilibrium from expert data in Markov Games. We show that a new quantity named the single policy deviation concentrability coefficient is unavoidable in the non-interactive imitation learning setting, and we provide an upper bound for behavioral cloning (BC) featuring such coefficient. BC exhibits substantial regret in games with high concentrability coefficient, leading us to utilize expert queries to develop and introduce two novel solution algorithms: MAIL-BRO and MURMAIL. The former employs a best response oracle and learns an $\varepsilon$-Nash equilibrium with $\mathcal{O}(\varepsilon^{-4})$ expert and oracle queries. The latter bypasses completely the best response oracle at the cost of a worse expert query complexity of order $\mathcal{O}(\varepsilon^{-8})$. Finally, we provide numerical evidence, confirming our theoretical findings.
format Preprint
id arxiv_https___arxiv_org_abs_2505_17610
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Learning Equilibria from Data: Provably Efficient Multi-Agent Imitation Learning
Freihaut, Till
Viano, Luca
Cevher, Volkan
Geist, Matthieu
Ramponi, Giorgia
Machine Learning
This paper provides the first expert sample complexity characterization for learning a Nash equilibrium from expert data in Markov Games. We show that a new quantity named the single policy deviation concentrability coefficient is unavoidable in the non-interactive imitation learning setting, and we provide an upper bound for behavioral cloning (BC) featuring such coefficient. BC exhibits substantial regret in games with high concentrability coefficient, leading us to utilize expert queries to develop and introduce two novel solution algorithms: MAIL-BRO and MURMAIL. The former employs a best response oracle and learns an $\varepsilon$-Nash equilibrium with $\mathcal{O}(\varepsilon^{-4})$ expert and oracle queries. The latter bypasses completely the best response oracle at the cost of a worse expert query complexity of order $\mathcal{O}(\varepsilon^{-8})$. Finally, we provide numerical evidence, confirming our theoretical findings.
title Learning Equilibria from Data: Provably Efficient Multi-Agent Imitation Learning
topic Machine Learning
url https://arxiv.org/abs/2505.17610