An iterative self-learning framework with multiple class prototypes corrects noisy labels and trains a ConvNet on both original and corrected labels, reporting strong accuracy on Clothing1M and Food101-N.
Training deep neural-networks using a noise adaptation layer
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Deep Self-Learning From Noisy Labels
An iterative self-learning framework with multiple class prototypes corrects noisy labels and trains a ConvNet on both original and corrected labels, reporting strong accuracy on Clothing1M and Food101-N.