Stable Deep Reinforcement Learning Method by Predicting Uncertainty in Rewards as a Subtask.

Kanata Suzuki, Tetsuya Ogata

Browse the full ICONIP paper archive.