Model-free reinforcement learning as mixture learning.

Nikos Vlassis, Marc Toussaint

Browse the full ICML paper archive.