Learning to Play in a Day: Faster Deep Reinforcement Learning by Optimality Tightening.
Frank S. HeYang LiuAlexander G. SchwingJian PengPublished in: ICLR (Poster) (2017)
Keyphrases
- reinforcement learning
- learning process
- learning algorithm
- state space
- learning systems
- learning problems
- action selection
- eligibility traces
- autonomous learning
- learning agent
- learning tasks
- transfer learning
- reinforcement learning algorithms
- background knowledge
- reinforcement learning methods
- prior knowledge
- evolutionary learning
- policy search
- machine learning