Reward Function and Initial Values: Better Choices for Accelerated Goal-Directed Reinforcement Learning.

Latitia Matignon, Guillaume J. Laurent, Nadine Le Fort-Piat

Browse the full ICANN paper archive.