Learning in Mean Field Games: the Fictitious Play

Abstract : Mean Field Game systems describe equilibrium configurations in differential games with infinitely many infinitesimal interacting agents. We introduce a learning procedure (similar to the Fictitious Play) for these games and show its convergence when the Mean Field Game is potential.
Document type :
Journal articles
Liste complète des métadonnées

https://hal.archives-ouvertes.fr/hal-01179503
Contributor : Pierre Cardaliaguet <>
Submitted on : Thursday, July 30, 2015 - 2:12:45 PM
Last modification on : Thursday, January 3, 2019 - 12:36:43 PM
Document(s) archivé(s) le : Saturday, October 31, 2015 - 10:34:17 AM

Files

LMFG20150730R.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01179503, version 2
  • ARXIV : 1507.06280

Collections

Citation

Pierre Cardaliaguet, Saeed Hadikhanloo. Learning in Mean Field Games: the Fictitious Play. ESAIM: Control, Optimisation and Calculus of Variations, EDP Sciences, 2017, 23 (2), pp.569-591. ⟨hal-01179503v2⟩

Share

Metrics

Record views

334

Files downloads

128