Training Classifiers that are Universally Robust to All Label Noise Levels.
Jingyi XuTony Q. S. QuekKai Fong Ernest ChongPublished in: CoRR (2021)
Keyphrases
- noise level
- training set
- noisy images
- test set
- noise estimation
- training samples
- training examples
- noise model
- class labels
- labeled images
- error rate
- noise reduction
- decision trees
- support vector
- training data
- classifier training
- gaussian noise
- learning algorithm
- sonar images
- training algorithm
- training process
- multiresolution
- feature selection