• Corpus ID: 7210677

Sharing Information Without Regret in Managed Stochastic Games

  title={Sharing Information Without Regret in Managed Stochastic Games},
  author={Michael J. Neely},
  • M. Neely
  • Published 30 December 2014
  • Economics
  • ArXiv
This paper considers information sharing in a multi-player repeated game. Every round, each player observes a subset of components of a random vector and then takes a control action. The utility earned by each player depends on the full random vector and on the actions of others. An example is a game where different rewards are placed over multiple locations, each player only knows the rewards in a subset of the locations, and players compete to collect the rewards. Sharing information can help… 

Figures from this paper


A Lyapunov optimization approach to repeated stochastic games
  • M. Neely
  • Economics
    2013 51st Annual Allerton Conference on Communication, Control, and Computing (Allerton)
  • 2013
A Lyapunov method is developed that solves the problem in an online max-weight fashion by selecting actions based on a set of time-varying weights, which does not require knowledge of the event probabilities.
Calibrated Learning and Correlated Equilibrium
For each correlated equilibrium there is some calibrated learning rule that the players can use which results in their playing this correlated equilibrium in the limit, and the statistical concept of a calibration is strongly related to the game theoretic concept of correlated equilibrium.
Equilibrium Points in N-Person Games.
  • J. Nash
  • Economics
    Proceedings of the National Academy of Sciences of the United States of America
  • 1950
One may define a concept of an n -person game in which each player has a finite set of pure strategies and in which a definite set of payments to the n players corresponds to each n -tuple of pure
Correlated Equilibrium in Stochastic Games
It is proved that any n-player stochastic game admits an autonomous correlated equilibrium payoff, when the game is positive and recursive, and a stationary correlation equilibrium payoff exists.
Non-cooperative games
Non-cooperative game models are described and game theoretic solution concepts are discussed, as well as refinements and relaxations of rationalizability and correlated equilibria.
A simple adaptive procedure leading to correlated equilibrium
We propose a new and simple adaptive procedure for playing a game: ‘‘regret-matching.’’ In this procedure, players may depart from their current play with probabilities that are proportional to
Strategically zero-sum games: The class of games whose completely mixed equilibria cannot be improved upon
In this paper we propose a new class of games, the “strategically zero-sum games,” which are characterized by a special payoff structure. We show that for a large body of correlation schemes which
Maximizing Queueing Network Utility Subject to Stability: Greedy Primal-Dual Algorithm
  • A. Stolyar
  • Computer Science
    Queueing Syst. Theory Appl.
  • 2005
A dynamic control algorithm, which is introduced, which the authors call Greedy Primal-Dual (GPD) algorithm, and prove its asymptotic optimality, and shows that it can accommodate a wide range of applications.
Conditional Universal Consistency
Players choose an action before learning an outcome chosen according to an unknown and history-dependent stochastic rule. Procedures that categorize outcomes, and use a randomized variation on
Asymptotic Properties of Proportional-Fair Sharing Algorithms
The Proportional Fair Scheduler of the Qualcomm High Data Rate (HDR) system and related algorithms are designed to deal with conflicts between fully exploiting the channel and being fair, and results of stochastic approximation are used to analyze the long term properties of this class.