Large Bandit Games

We study a multi-player one-arm bandit game: for infinitely many stages, players choose between playing a risky action or dropping out irreversibly to a safe action. Each player observe his own payoffs and other players' actions. We study equilibria of the game when the number of players gets large. We argue that either limit equilibrium exhibits aggregate randomness, or players behaviours leads to a swift determination of the value of the risky action. The conditions under which one or the other case arises are provided.

Data and Resources

Additional Info

Field Value
Source https://hal.science/hal-00562257
Author Salomon, Antoine
Maintainer CCSD
Last Updated November 6, 2026, 23:15 (UTC)
Created November 6, 2026, 23:15 (UTC)
Identifier hal-00562257
Language en
Rights https://about.hal.science/hal-authorisation-v1/
contributor imagine [Marne-la-Vallée] ; Laboratoire d'Informatique Gaspard-Monge (LIGM) ; Université Paris-Est Marne-la-Vallée (UPEM)-École nationale des ponts et chaussées (ENPC)-ESIEE Paris-Fédération de Recherche Bézout (BEZOUT) ; Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS)-Université Paris-Est Marne-la-Vallée (UPEM)-École nationale des ponts et chaussées (ENPC)-ESIEE Paris-Fédération de Recherche Bézout (BEZOUT) ; Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS)-Centre National de la Recherche Scientifique (CNRS)-Centre Scientifique et Technique du Bâtiment (CSTB)
creator Salomon, Antoine
date 2010-10-01T00:00:00
harvest_object_id bb1c6906-27fd-4e0d-9d88-70ac92877510
harvest_source_id 3374d638-d20b-4672-ba96-a23232d55657
harvest_source_title test moissonnage SELUNE
metadata_modified 2026-06-06T00:00:00
set_spec type:UNDEFINED