eprintid: 4179 rev_number: 23 eprint_status: archive userid: 351 dir: disk0/00/00/41/79 datestamp: 2016-01-15 02:04:39 lastmod: 2021-08-27 17:14:55 status_changed: 2016-01-15 02:04:39 type: monograph metadata_visibility: show item_issues_count: 3 creators_name: Kaniovski, Y.M. creators_name: Young, H.P. creators_id: AL0815 creators_id: AL1397 title: Learning Dynamics in Games with Stochastic Perturbations ispublished: pub internal_subjects: iis_ecn internal_subjects: iis_frc internal_subjects: iis_mnt internal_subjects: iis_met divisions: prog_ted abstract: Consider a game that is played repeatedly by two populations of agents. In fictitious play, agents learn by choosing best replies to the frequency distribution of actions taken by the other side. We consider a more general class of learning processes in which agents' choices are perturbed by incomplete information about what the other side has done, variability in their payoffs, and unexplained trembles. These perturbed best reply dynamics define a non-stationary Markov process on an infinite state space. We show that for 2x2 games it converges with probability one to a neighborhood of the stable Nash equilibria, whether pure or mixed. This generalizes a result of Fudenberg and Kreps, who demonstrate convergence when the game has a unique mixed equilibrium. date: 1994-12 date_type: published publisher: WP-94-030 iiasapubid: WP-94-030 price: 10 creators_browse_id: 2009 creators_browse_id: 2539 full_text_status: public monograph_type: working_paper place_of_pub: IIASA, Laxenburg, Austria pages: 35 coversheets_dirty: FALSE fp7_type: info:eu-repo/semantics/book citation: Kaniovski, Y.M. & Young, H.P. (1994). Learning Dynamics in Games with Stochastic Perturbations. IIASA Working Paper. IIASA, Laxenburg, Austria: WP-94-030 document_url: https://pure.iiasa.ac.at/id/eprint/4179/1/WP-94-030.pdf