Best Reply Dynamics for Scoring Rules
نویسندگان
چکیده
We consider best-reply dynamics for voting games in which all players are strategic and no coalitions are formed. We study the class of scoring rules, show convergence of a suitably restricted version for the plurality and veto rules, and failure of convergence for other rules including k-approval and Borda. In particular, for 3 candidates convergence fails for all rules other than plurality and veto. We give a unified proof for the convergence of these two rules. Our proofs in the case of plurality improve the known bound on convergence, and the other convergence results are new.
منابع مشابه
On the Evolution of Optimizing Behavior*
We investigate whether replicator dynamics justify “survival of the fittest” when players inherit pure strategies, mixed strategies, or rules for choosing strategies. A strategy that never is a best reply can survive replicator dynamics if only pure strategies are heritable, but only rationalizable strategies survive if mixed strategies are heritable. Now suppose agents can inherit the rule of ...
متن کاملRefined best reply correspondence and dynamics
We call a correspondence, defined on the set of mixed strategy profiles, a generalized best reply correspondence if it (1) has a product structure, (2) is upper hemi–continuous, (3) always includes a best reply to any mixed strategy profile, and (4) is convexand closed-valued. For each generalized best reply correspondence, we define a generalized best reply dynamics as a differential inclusion...
متن کاملBest-Reply Mechanisms
In many real-world settings, strategic agents are instructed to follow best-reply dynamics (economic markets, Internet protocols, etc.). Such settings give rise to many questions: Will best-reply dynamics “converge” or go on indefinitely? Will they converge quickly? What happens in distributed and asynchronous settings in which communication between players may be delayed or lost? Is it in the ...
متن کاملAsynchronous Best-Reply Dynamics
In many real-world settings (e.g., interdomain routing in the Internet) strategic agents are instructed to follow best-reply dynamics in asynchronous environments. In such settings players learn of each other’s actions via update messages that can be delayed or even lost. In particular, several players might update their actions simultaneously, or make choices based on outdated information. In ...
متن کاملRefined best-response correspondence and dynamics
We study a natural (and, in a well-defined sense, minimal) refinement of the best-reply correspondence. Its fixed points, notions of rationalizability and CURB sets based on this correspondence are defined and characterized. Finally minimally asymptotically stable faces under the induced refined best-reply dynamics are characterized.
متن کامل