Reinforcement Learning for Relational MDPs

In this paper we present a new method for reinforcement learning in relational domains. A logical language is employed to abstract over states and actions, thereby decreasing the size of the state-action space significantly. A probabilistic transition model of the abstracted Markov-Decision-Process is estimated to to speed-up learning. We present theoretical and experimental analysis of our new representation. Some insights concerning the problems and opportunities of logical representations for reinforcement learning are obtained in the context of a growing interest in the use of abstraction in reinforcement learning contexts.