Active Learning on Heterogeneous Information Networks: A Multi-armed Bandit Approach.
Doris XinAhmed El-KishkyDe LiaoBrandon NorickJiawei HanPublished in: ICDM (2018)
Keyphrases
- heterogeneous information networks
- active learning
- multi armed bandit
- heterogeneous networks
- information networks
- multi armed bandits
- reinforcement learning
- multiple types
- learning algorithm
- relevance feedback
- topic models
- training examples
- data objects
- random walk
- supervised learning
- real world
- machine learning
- training set
- learning process
- semi supervised
- regret bounds
- unlabeled data
- social networks
- link prediction
- cost sensitive
- knn
- upper bound
- generalization error
- network analysis
- bandit problems