Probabilistic End-to-end Noise Correction for Learning with Noisy Labels

Abstract

Deep learning has achieved excellent performance in various computer vision tasks, but requires a lot of training examples with clean labels. It is easy to collect a dataset with noisy labels, but such noise makes networks overfit seriously and accuracies drop dramatically. To address this problem, we propose an end-to-end framework called PENCIL, which can update both network parameters and label estimations as label distributions. PENCIL is independent of the backbone network structure and does not need an auxiliary clean dataset or prior information about noise, thus it is more general and robust than existing methods and is easy to apply. PENCIL outperforms previous state-of-the-art methods by large margins on both synthetic and real-world datasets with different noise types and noise rates. Experiments show that PENCIL is robust on clean datasets, too.

CVPR 2019

References in corpus (1)

• Training Convolutional Networks with Noisy Labels

Cited by in corpus (8)

• DivideMix: Learning with Noisy Labels as Semi-supervised Learning
• Combating noisy labels by agreement: A joint training method with co-regularization
• MoPro: Webly Supervised Learning with Momentum Prototypes
• Multi-Class Classification from Noisy-Similarity-Labeled Data
• Network Cooperation with Progressive Disambiguation for Partial Label Learning
• A Tale Of Two Long Tails
• Do We Really Need Gold Samples for Sample Weighting Under Label Noise?
• Batch Inverse-Variance Weighting: Deep Heteroscedastic Regression