Switching between different state representations in reinforcement learning

This paper proposes a reinforcement learning architecture containing multiple “experts”, each of which is a specialist in a different region in the overall state space. The central idea is that the different experts use qualitatively different (but sufficiently Markov) state representations, each of which captures different information regarding the true underlying world state, and which for that reason is suitable for a different part of the state space. The experts themselves learn to switch to another state representation (other expert) by having switching actions. Value functions can be learned using standard reinforcement learning algorithms. This architecture has important advantages in RL problems that have large state spaces or where a sensor system must inherently choose between mutually exclusive state representations. Experiments in a small, proof-of-principle experiment as well as a larger, more realistic experiment illustrate the validity of this approach.