ProtoCon: Pseudo-label Refinement via Online Clustering and Prototypical Consistency for Efficient Semi-supervised Learning.
Islam NassarMunawar HayatEhsan AbbasnejadHamid RezatofighiGholamreza HaffariPublished in: CoRR (2023)
Keyphrases
- semi supervised learning
- semi supervised
- unsupervised learning
- label propagation
- labeled data
- unlabeled data
- supervised learning
- semi supervised clustering
- dirichlet process mixture models
- multi label classification
- semi supervised classification
- multi instance learning
- machine learning
- active learning
- data points
- graph based semi supervised learning
- graph construction
- partially labeled
- clustering algorithm
- co training
- spectral clustering
- learning models
- manifold regularization
- clustering method
- semi supervised learning methods
- pair wise constraints
- text categorization
- k means
- training data
- data clustering
- labeled examples
- transfer learning
- nonnegative matrix factorization
- multi label
- class labels
- text classification
- opinion detection