Stochastic learning models provide sharp predictions about equilibrium selection when the noise level of the learning process is taken to zero.  The difficulty is that, when the noise is extremely small, it can take an extremely long time for a large population to reach the stochastically...

The diffusion of an innovation can be represented by a process in which agents choose perturbed best responses to what their neighbors are currently doing.  Diffusion is said to be fast if the expected waiting time until the innovation spreads widely is bounded above independently of the size...

We propose a simple payoff-based learning rule that is completely decentralized, and that leads to an efficient configuaration of actions in any n-person finite strategic-form game with generic payoffs.  The algorithm follows the theme of exploration versus exploitation and is hence stochastic...

An individual's learning rule is completely uncoupled if it does not depend on the actions or payoffs of anyone else.  We propose a variant of log linear learning that is completely uncoupled and that selects an efficient pure Nash equilibrium in all generic n-person games that possess at least...

This paper demonstrates that inertia driven by switching costs leads to more rapid evolution in a class of games that includes m x m pure coordination games. Under the best-response dynamic and a fixed rate of mutation, the expected waiting time to reach long-run equilibrium is of lower order in...

In von Neumann and Morgenstern's sample model of poker, equilibrium has the first player bet with high and low hands, and check with intermediate hands.  The second player then calls if his hand is sufficiently high.  Betting by the low hands is interpreted as bluffing, and is a pure...

Many interactive environments can be represented as games, but they are so large and complex that individual players are in the dark about others' actions and the payoff structure.  This paper analyzes learning behavior in such 'black box' environments, where players' only source of information...

The World's nations have yet to reach a truly effective treaty to control the emission of greenhouse gases.  The importance of compatibility with private incentives of individual countries has been acknowledged (at least by game theorists) in designing climate policies for the post-Kyoto...

If players learn to play an infinitely repeated game using Bayesian learning, it is known that their strategies eventually approximate Nash equilibria of the repeated game under an absolute-continuity assumption on their prior beliefs.  We suppose here that Bayesian learners do not start with...

Social norms are patterns of behavior that are self-enforcing at the group level: everyone wants to conform when they expect everyone else to conform.  There are multiple mechanisms that sustain social norms, including a desire to coordinate, fear of being sanctioned, signaling membership in...