Reinforcement learning with value advice.

Mayank Daswani, Peter Sunehag, Marcus Hutter

Browse the full ACML paper archive.