Predictive Representations of State

We show that states of a dynamical system can be usefully represented by multi-step, action-conditional predictions of future observations. State representations that are grounded in data in this way may be easier to learn, generalize better, and be less dependent on accurate prior models than, for example, POMDP state representations. Building on prior work by Jaeger and by Rivest and Schapire, in this paper we compare and contrast a linear specialization of the predictive approach with the state representations used in POMDPs and in k-order Markov models. Ours is the first specific formulation of the predictive idea that includes both stochasticity and actions (controls). We show that any system has a linear predictive state representation with number of predictions less than or equal to the number of states in its minimal POMDP model. In predicting or controlling a sequence of observations, the concepts of state and state estimation inevitably arise. There have been two dominant ap...

• Type: Conference
• Year: 2001
• Where: NIPS
• Authors: Michael L. Littman, Richard S. Sutton, Satinder P. Singh
• Keywords: K-order Markov, NIPS 2001, NIPS 2007, POMDP State Representations, State Representation