Generalization in reinforcement learning: Safely approximating the value function (1995)

by J A Boyan, A W Moore
Venue:In Advances in Neural Information Processing Systems 7 (NIPS 7