Regularizing Feature Distribution Using Sliced Wasserstein Distance for Semi-supervised Learning.
Jinhyung KimChanho LeeJunmo KimPublished in: MIWAI (2018)
Keyphrases
- semi supervised learning
- unlabeled data
- semi supervised
- labeled data
- semi supervised classification
- supervised learning
- unsupervised learning
- co training
- active learning
- machine learning
- labeled examples
- distance measure
- training data
- improve the classification accuracy
- manifold regularization
- graph based semi supervised learning
- learning problems
- transfer learning
- distance function
- learning models
- label propagation
- semi supervised learning methods
- semi supervised learning algorithms
- labeled and unlabeled data
- data mining
- learning algorithm
- feature vectors
- training and test data
- regularization framework
- graph construction
- image features
- text classification
- data distribution
- euclidean distance
- class labels
- feature set
- multiple features
- domain adaptation