Near-Optimal Reinforcement Learning in Polynominal Time.

Michael J. Kearns, Satinder Singh

Browse the full ICML paper archive.