Advantage based value iteration for Markov decision processes with unknown rewards.

Pegah Alizadeh, Yann Chevaleyre, Franois Lvy

Browse the full IJCNN paper archive.