Reinforcement learning with fuzzy evaluative feedback for a biped robot | IEEE Conference Publication | IEEE Xplore