Networks of learning automata and limiting games.

Katja Verbeeck, Peter Vrancx, Ann Nowe, Karl Tuyls (Editor), Steven De Jong (Editor), Marc Ponsen (Editor), Katja Verbeeck (Editor)

Research output: Contribution to journalConference paper

Abstract

Abstract. Learning Automata (LA) were recently shown to be valuable
tools for designing Multi-Agent Reinforcement Learning algorithms. One
of the principal contributions of LA theory is that a set of decentralized,
independent learning automata is able to control a finite Markov Chain
with unknown transition probabilities and rewards. This result was re-
cently extended to Markov Games and analyzed with the use of limiting
games. In this paper we continue this analysis but we assume here that
our agents are fully ignorant about the other agents in the environment,
i.e. they can only observe themselves; they don't know how many other
agents are present in the environment, the actions these other agents
took nor the rewards they received for this, nor the location they
occupy in the state space. We prove that in Markov Games, where agents
have this limited type of observability, a network of independent LA is
still able to converge to an equilibrium point of the underlying limiting
game, provided a common ergodic assumption and provided the agents
do not interfere each other's transition probabilities.
Original languageEnglish
Pages (from-to)171-183
Number of pages13
JournalTechnical report in mathematics and computer science
Publication statusPublished - 2007
EventSeventh European Symposium on Adaptive and Learning Agents and Multi-Agent Systems (ALAMAS'07) - Maastricht University, Maastricht, Netherlands
Duration: 2 Apr 20073 Apr 2007

Keywords

  • learning automata
  • multi-agent
  • partial observability
  • limiting game

Fingerprint

Dive into the research topics of 'Networks of learning automata and limiting games.'. Together they form a unique fingerprint.

Cite this