USB: A Unified Semi-supervised Learning Benchmark for Classification.
Yidong WangHao ChenYue FanWang SunRan TaoWenxin HouRenjie WangLinyi YangZhi ZhouLan-Zhe GuoHeli QiZhen WuYu-Feng LiSatoshi NakamuraWei YeMarios SavvidesBhiksha RajTakahiro ShinozakiBernt SchieleJindong WangXing XieYue ZhangPublished in: NeurIPS (2022)
Keyphrases
- semi supervised learning
- supervised learning
- semi supervised classification
- improve the classification accuracy
- unlabeled data
- co training
- semi supervised
- machine learning
- transductive support vector machine
- unsupervised learning
- labeled and unlabeled data
- semi supervised learning algorithms
- uci datasets
- labeled data
- classification accuracy
- training data
- class labels
- supervised learning algorithms
- text categorization
- feature extraction
- manifold regularization
- graph based semi supervised learning
- training set
- decision boundary
- instance selection
- active learning
- pattern recognition
- multi view
- transfer learning
- learning models
- supervised classification
- labeled examples
- labeled instances
- unlabeled samples
- decision trees
- discriminative learning
- label propagation
- semi supervised learning methods
- feature space
- pairwise
- real world
- graph construction
- pairwise constraints
- training samples
- support vector
- principal component analysis
- high dimensional