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