Fast active learning for pure exploration
in reinforcement learning
Pierre Ménard 1 Omar Darwiche Domingues 2 Emilie Kaufmann 2 3 Anders Jonsson 4 Edouard Leurent 2
Michal Valko 2 3 5
Abstract how to explore efficiently. In particular we wish to com-
pute near-optimal policies using the least possible amount
Realistic environments often provide agents with
...


雷达卡




京公网安备 11010802022788号







