Maximum Entropy Semi-Supervised Inverse Reinforcement Learning.
Julien AudiffrenMichal ValkoAlessandro LazaricMohammad GhavamzadehPublished in: IJCAI (2015)
Keyphrases
- maximum entropy
- inverse reinforcement learning
- semi supervised
- preference elicitation
- semi supervised learning
- maximum entropy principle
- labeled data
- markov models
- unlabeled data
- unsupervised learning
- supervised learning
- active learning
- pairwise
- reward function
- multi view
- co training
- conditional random fields
- metric learning
- generative model
- pairwise constraints
- semi supervised clustering
- gaussian process
- utility function
- similarity measure
- temporal difference
- kernel logistic regression