Showing papers by "Andreu Mas-Colell published in 2001"
TL;DR: In this article, the authors characterize a class of simple adaptive strategies, in the repeated play of a game, having the Hannanconsistency property: in the long run, the player is guaranteed an average payoff as large as the best-reply payoff to the empirical distribution of play of the other players; i.e., there is no regret.
233 citations
01 Jan 2001
TL;DR: The authors consider repeated games where at any period each player knows only his set of actions and the stream of payoffs that he has received in the past, and present an adaptive procedure for play called modified-regret-matching, which is interpretable as a stimulus-response or reinforcement procedure.
Abstract: We consider repeated games where at any period each player knows only his set of actions and the stream of payoffs that he has received in the past. He knows neither his own payoff function, nor the characteristics of the other players (how many there are, their strategies and payoffs). In this context, we present an adaptive procedure for play called “modified-regret-matching” — which is interpretable as a stimulus-response or reinforcement procedure, and which has the property that any limit point of the empirical distribution of play is a correlated equilibrium of the stage game.
166 citations
01 Jan 2001
TL;DR: In this paper, Hart and Mas-Colell [11] showed that one needs to replace 1/(m i − 1) by 1/m i in Formula (7) by p i t+1+1 (k) := 1 − δ t γ min 1 µ Q i t (j, k), 1 m i + δt γ
Abstract: In Hart and Mas-Colell [2001], one needs to replace 1/(m i − 1) by 1/m i in Formula (7); that is, p i t+1 (k) := 1 − δ t γ min 1 µ Q i t (j, k) , 1 m i + δ t γ
2 citations