Learning to Shape Rewards Using a Game of Two Partners.
David MguniTaher JafferjeeJianhong WangNicolas Perez NievesWenbin SongFeifei TongMatthew E. TaylorTianpei YangZipeng DaiHui ChenJiangcheng ZhuKun ShaoJun WangYaodong YangPublished in: AAAI (2023)
Keyphrases
- learning algorithm
- reinforcement learning
- prior knowledge
- computer games
- supervised learning
- knowledge acquisition
- role playing
- shape features
- shape analysis
- learning tasks
- markov decision processes
- learning systems
- learning process
- online learning
- state space
- evaluation function
- active learning
- board game
- two player games
- machine learning