Cooperation in stochastic games through communication - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2005

Cooperation in stochastic games through communication

Raghav Aras
  • Function : Author
  • PersonId : 830439
Alain Dutech

Abstract

We describe a process of reinforcement learning in two-agent general-sum stochastic games under imperfect observability of moves and payoffs. In practice, it is known that using naive Q-learning, agents can learn equilibrium policies under the discounted reward criterion although these may be arbitrarily worse for both the agents than a non-equilibrium policy, in the absence of global optima. We aim for Pareto-efficiency in policies, in which agents enjoy higher payoffs than in an equilibrium and show agents may employ naive Q-learning with the addition of communication and a payoff interpretation rule, to achieve this. In principle, our objective is to shift the focus of the learning from equilibria (to which solipsistic algorithms converge) to non-equilibria by transforming the latter to equilibria.
Fichier principal
Vignette du fichier
p525-aras.pdf (58.6 Ko) Télécharger le fichier

Dates and versions

inria-00000208 , version 1 (13-09-2005)

Identifiers

Cite

Raghav Aras, Alain Dutech, François Charpillet. Cooperation in stochastic games through communication. 4th International Joint Conference on Autonomous Agents and Multiagent Systems - AAMAS'05, Jul 2005, Utrecht/ The Netherlands, pp.1197 - 1198, ⟨10.1145/1082473.1082691⟩. ⟨inria-00000208⟩
172 View
156 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More