Actively learning equilibria in Nash games with misleading information

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Franci, Barbara, Fabiani, Filippo, Bemporad, Alberto
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915276062195712
author Franci, Barbara
Fabiani, Filippo
Bemporad, Alberto
author_facet Franci, Barbara
Fabiani, Filippo
Bemporad, Alberto
contents We develop a scheme based on active learning to compute equilibria in a generalized Nash equilibrium problem (GNEP). Specifically, an external observer (or entity), with little knowledge on the multi-agent process at hand, collects sensible data by probing the agents' best-response (BR) mappings, which are then used to recursively update local parametric estimates of these mappings. Unlike [1], we consider the realistic case in which the agents share corrupted information with the external entity for, e.g., protecting their privacy. Inspired by a popular approach in stochastic optimization, we endow the external observer with an inexact proximal scheme for updating the local BR proxies. This technique will prove key to establishing the convergence of our scheme under standard assumptions, thereby enabling the external observer to predict an equilibrium strategy even when relying on masked information.
format Preprint
id arxiv_https___arxiv_org_abs_2503_13167
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Actively learning equilibria in Nash games with misleading information
Franci, Barbara
Fabiani, Filippo
Bemporad, Alberto
Optimization and Control
We develop a scheme based on active learning to compute equilibria in a generalized Nash equilibrium problem (GNEP). Specifically, an external observer (or entity), with little knowledge on the multi-agent process at hand, collects sensible data by probing the agents' best-response (BR) mappings, which are then used to recursively update local parametric estimates of these mappings. Unlike [1], we consider the realistic case in which the agents share corrupted information with the external entity for, e.g., protecting their privacy. Inspired by a popular approach in stochastic optimization, we endow the external observer with an inexact proximal scheme for updating the local BR proxies. This technique will prove key to establishing the convergence of our scheme under standard assumptions, thereby enabling the external observer to predict an equilibrium strategy even when relying on masked information.
title Actively learning equilibria in Nash games with misleading information
topic Optimization and Control
url https://arxiv.org/abs/2503.13167