Semi-Supervised Representation Learning via Triplet Loss Based on Explicit Class Ratio of Unlabeled Data.
Kazuhiko MurasakiShingo AndoJun ShimamuraPublished in: IEICE Trans. Inf. Syst. (2022)
Keyphrases
- semi supervised
- unlabeled data
- labeled and unlabeled data
- labeled data
- supervised learning
- semi supervised learning
- active learning
- co training
- positive examples
- learning algorithm
- class labels
- unsupervised learning
- partially labeled data
- semi supervised classification
- label information
- multi view learning
- transductive learning
- labeled examples
- supervised and semi supervised
- training data
- prior knowledge
- learning process
- domain adaptation
- partially labeled
- learning tasks
- labeled samples
- weakly supervised
- sample selection bias
- training examples
- positive and unlabeled examples
- pairwise
- label propagation
- text classification
- semi supervised learning algorithms
- fully supervised
- machine learning
- labeling effort
- supervised learning algorithms
- pairwise constraints
- decision boundary
- learning problems
- data points
- semi supervised clustering
- kernel learning
- labeled instances
- unlabeled instances
- learning models
- dimensionality reduction