Manifold Embeddings for Model-Based Reinforcement Learning under Partial Observability

Keith Bush, Joelle Pineau
2009 Neural Information Processing Systems  
Interesting real-world datasets often exhibit nonlinear, noisy, continuous-valued states that are unexplorable, are poorly described by first principles, and are only partially observable. If partial observability can be overcome, these constraints suggest the use of model-based reinforcement learning. We experiment with manifold embeddings to reconstruct the observable state-space in the context of offline, model-based reinforcement learning. We demonstrate that the embedding of a system can
more » ... ange as a result of learning, and we argue that the best performing embeddings well-represent the dynamics of both the uncontrolled and adaptively controlled system. We apply this approach to learn a neurostimulation policy that suppresses epileptic seizures on animal brain slices.
dblp:conf/nips/BushP09 fatcat:orqjthixyvf4lkkwbnknii6v6i