Improving MACS thanks to a comparison with 2TBNs

Thierry Gourdin (1) , Olivier Sigaud (1) , Pierre-Henri Wuillemin (2)

1 Animatlab
2 DECISION

Factored Markov Decision Processes is the theoretical framework underlying multi-step Learning Classifier Systems research. This framework is mostly used in the context of Two-stage Bayes Networks, a subset of Bayes Networks. In this paper, we compare the Learning Classifier Systems approach and the Bayes Networks approach to factored Markov Decision Problems. More specifically, we focus on a comparison between MACS, an Anticipatory Learning Classifier System, and Structured Policy Iteration, a general planning algorithm used in the context of Two-stage Bayes Networks. From that comparison, we define a new algorithm resulting from the adaptation of Structured Policy Iteration to the context of MACS. We conclude by calling for a closer communication between both research communities.

Domaines: Informatique [cs]

Thierry Gourdin, Olivier Sigaud, Pierre-Henri Wuillemin. Improving MACS thanks to a comparison with 2TBNs. GECCO 2004 - Genetic and Evolutionary Computation Conference, Jun 2004, Seattle, WA, United States. pp.810-823, ⟨10.1007/978-3-540-24855-2_95⟩. ⟨hal-01501406⟩