Self-Filtering: A Noise-Aware Sample Selection for Label Noise with Confidence Penalization.
Qi WeiHaoliang SunXiankai LuYilong YinPublished in: ECCV (30) (2022)
Keyphrases
- label noise
- active learning
- sample selection
- training data
- generalization error
- noise model
- machine learning
- noise tolerant
- training examples
- learning algorithm
- semi supervised
- supervised learning
- noisy data
- noise level
- learning process
- labeled data
- multi class
- classification accuracy
- missing data
- cross validation
- training set
- support vector
- feature extraction
- feature selection