Tracking value function dynamics to improve reinforcement learning with piecewise linear function approximation.

Chee Wee Phua, Robert Fitch

Browse the full ICML paper archive.