Logarithmic regret of exploration in average reward Markov decision processes.

Victor Boone, Bruno Gaujal

Browse the full COLT paper archive.