ConferenceAAAI Conference on Artificial Intelligence2016Featured
Deep Reinforcement Learning with Double Q-Learning
Hado van Hasselt, Arthur Guez, David Silver
This paper shows that DQN overestimates action values in Atari games and proposes a Double DQN algorithm that reduces overestimation and improves performance.
9.3kMar 2, 2016Reinforcement LearningNeural Networks
arXiv