Learning to Combat Noisy Labels via Classification Margins
arXiv:2102.00751
Abstract
A deep neural network trained on noisy labels is known to quickly lose its power to discriminate clean instances from noisy ones. After the early learning phase has ended, the network memorizes the noisy instances, which leads to a significant degradation in its generalization performance. To resolve this issue, we propose MARVEL (MARgins Via Early Learning), a new robust learning method where the memorization of the noisy instances is curbed. We propose a new test statistic that tracks the goodness of "fit" of every instance based on the epoch-history of its classification margins. If its classification margin is small in a sequence of consecutive learning epochs, that instance is declared noisy and the network abandons learning on it. Consequently, the network first flags a possibly noisy instance, and then waits to see if learning on that instance can be improved and if not, the network learns with confidence that this instance can be safely abandoned. We also propose MARVEL+, where arduous instances can be upweighted, enabling the network to focus and improve its learning on them and consequently its generalization. Experimental results on benchmark datasets with synthetic label noise and real-world datasets show that MARVEL outperforms other baselines consistently across different noise levels, with a significantly larger margin under asymmetric noise.
arXiv comment23 pages, 9 sets of figures
References in corpus (19)
- Learning to Reweight Examples for Robust Deep Learning
- Classification with Noisy Labels by Importance Reweighting
- Training Deep Neural Networks on Noisy Labels with Bootstrapping
- A Closer Look at Memorization in Deep Networks
- WebVision Database: Visual Learning and Understanding from Web Data
- Dimensionality-Driven Learning with Noisy Labels
- Making Risk Minimization Tolerant to Label Noise
- Unsupervised Label Noise Modeling and Loss Correction
- Gradient Descent with Early Stopping is Provably Robust to Label Noise for Overparameterized Neural Networks
- How does Disagreement Help Generalization against Label Corruption?
- Robust Loss Functions under Label Noise for Deep Neural Networks
- Curriculum Loss: Robust Learning and Generalization against Label Corruption
- Combating Label Noise in Deep Learning Using Abstention
- Understanding and Utilizing Deep Neural Networks Trained with Noisy Labels
- L_DMI: An Information-theoretic Noise-robust Loss Function
- Learning Not to Learn in the Presence of Noisy Labels
- How does Early Stopping Help Generalization against Label Noise?
- Learning Adaptive Loss for Robust Learning with Noisy Labels
- Coresets for Robust Training of Neural Networks against Noisy Labels