Using random complementary labels as negative examples, then selectively applying positive learning to high-confidence samples, gives state-of-the-art accuracy on image classification with noisy labels.
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.LG 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
NLNL: Negative Learning for Noisy Labels
Using random complementary labels as negative examples, then selectively applying positive learning to high-confidence samples, gives state-of-the-art accuracy on image classification with noisy labels.