SemiNLL: A Framework of Noisy-Label Learning by Semi-Supervised Learning.
Zhuowei WangJing JiangBo HanLei FengBo AnGang NiuGuodong LongPublished in: CoRR (2020)
Keyphrases
- semi supervised learning
- supervised learning
- multi view learning
- unsupervised learning
- boosting framework
- learning frameworks
- labeled and unlabeled data
- semi supervised
- active learning
- learning problems
- unlabeled data
- co training
- labeled data
- learning process
- label propagation
- semi supervised learning setting
- regularization framework
- manifold regularization
- graph transduction
- partially labeled
- covariate shift
- learning models
- learning tasks
- machine learning
- learning algorithm
- data mining
- class labels
- multiple instance learning
- feature selection
- loss function
- data sets
- reinforcement learning
- inductive inference
- domain adaptation
- semi supervised classification
- training data
- pairwise
- expectation maximization
- training set
- prior knowledge
- kernel learning
- positive examples
- multi task
- transfer learning