Active Learning for Reward Estimation in Inverse Reinforcement Learning.
Manuel LopesFrancisco S. MeloLuis MontesanoPublished in: ECML/PKDD (2) (2009)
Keyphrases
- inverse reinforcement learning
- active learning
- partially observable environments
- bayesian nonparametric
- reward function
- preference elicitation
- random sampling
- temporal difference
- artificial intelligence
- learning algorithm
- multi attribute
- special case
- partially observable
- reinforcement learning
- machine learning
- semi supervised
- reinforcement learning algorithms
- training data