Part of Advances in Neural Information Processing Systems 25 (NIPS 2012)
Ronald Ortner, Daniil Ryabko
We derive sublinear regret bounds for undiscounted reinforcement learning in continuous state space. The proposed algorithm combines state aggregation with the use of upper confidence bounds for implementing optimism in the face of uncertainty. Beside the existence of an optimal policy which satisfies the Poisson equation, the only assumptions made are Hoelder continuity of rewards and transition probabilities.