In this paper, we introduce a novel form of value function, Q(s,s′), that expresses the utility of transitioning from a state s to a neighboring state s′ and then acting optimally thereafter. In order to derive an optimal policy, we develop a forward dynamics model that learns to make next-state predictions that maximize this value. This formulation decouples actions from values while still learning off-policy. We highlight the benefits of this approach in terms of value function transfer, learning within redundant action spaces, and learning off-policy from state observations generated by sub-optimal or completely random policies. Code and videos are available at this http URL.
37th International Conference on Machine Learning (ICML), 2020