This study examines how each player chooses her/his optimal action in "normal-form games with unawareness" by applying a "discovery process" to them. We show that if each player implements a best response to the opponents' immediately preceding plays, then any discovery process converges to a...

This paper studies the reinforcement learning of Erev and Roth with foregone payoff information in normal form games: players observe not only the realised payoffs but also foregone payoffs, the ones which they could have obtained if they had chosen the other actions. We provide conditions under...

In the literature of psychology and economics it is frequently observed that individuals tend to imitate similar individuals. A fundamental question is whether the outcome of such imitation can be consistent with self-interested behaviour. We propose that this consistency requires the existence...

This paper provides a definition of epistemic stability of sets of strategy profiles, and uses it to characterize variants of curb sets in finite games, including the set of rationalizable strategies and minimal curb sets. -- Epistemic game theory ; epistemic stability ; rationalizability ;...

We study experimentally whether heterogeneity of behavior in the Centipede game can be interpreted as the result of a learning process of individuals with different preference types (more and less pro-social) and coarse information regarding the opponent's past behavior. We manipulate the...

We examine rational learning among expert chess players and how they update their beliefs in repeated games with the same opponent. We present a model that explains how equilibrium play is affected when players change their choice of strategy when receiving additional information from each...

We study experimentally whether heterogeneity of behavior in the Centipede game can be interpreted as the result of a learning process of individuals with different preference types (more and less pro-social) and coarse information regarding the opponent's past behavior. We manipulate the...

A central result in the rational learning literature is that if the true measure is absolutely continuous with respect to the beliefs then, given enough data, the updated beliefs merge with the true distribution. In this paper, we show that, under absolute continuity, weak merging occurs fast...