LongReMix: Robust Learning with High Confidence Samples in a Noisy Label Environment

Abstract

Deep neural network models are robust to a limited amount of label noise, but their ability to memorise noisy labels in high noise rate problems is still an open issue. The most competitive noisy-label learning algorithms rely on a 2-stage process comprising an unsupervised learning to classify training samples as clean or noisy, followed by a semi-supervised learning that minimises the empirical vicinal risk (EVR) using a labelled set formed by samples classified as clean, and an unlabelled set with samples classified as noisy. In this paper, we hypothesise that the generalisation of such 2-stage noisy-label learning methods depends on the precision of the unsupervised classifier and the size of the training set to minimise the EVR. We empirically validate these two hypotheses and propose the new 2-stage noisy-label training algorithm LongReMix. We test LongReMix on the noisy-label benchmarks CIFAR-10, CIFAR-100, WebVision, Clothing1M, and Food101-N. The results show that our LongReMix generalises better than competing approaches, particularly in high label noise problems. Furthermore, our approach achieves state-of-the-art performance in most datasets. The code is available at https://github.com/filipe-research/LongReMix.

arXiv commentPublished at Pattern Recognition 2022

References in corpus (14)

• A Survey on Deep Learning in Medical Image Analysis
• mixup: Beyond Empirical Risk Minimization
• Understanding deep learning requires rethinking generalization
• Learning to Reweight Examples for Robust Deep Learning
• DivideMix: Learning with Noisy Labels as Semi-supervised Learning
• WebVision Database: Visual Learning and Understanding from Web Data
• Dimensionality-Driven Learning with Noisy Labels
• MAIN: Multihead-Attention Imputation Networks
• How does Disagreement Help Generalization against Label Corruption?
• Beyond Synthetic Noise: Deep Learning on Controlled Noisy Labels
• Combating Label Noise in Deep Learning Using Abstention
• Robust Inference via Generative Classifiers for Handling Noisy Labels
• IMAE for Noise-Robust Learning: Mean Absolute Error Does Not Treat Examples Equally and Gradient Magnitude's Variance Matters
• Generalization Bounds for Vicinal Risk Minimization Principle

Cited by in corpus (5)

• FINE Samples for Learning with Noisy Labels
• SSR: An Efficient and Robust Framework for Learning with Unknown Label Noise
• Robust Remote Sensing Scene Classification with Multi-View Voting and Entropy Ranking
• Sample Prior Guided Robust Model Learning to Suppress Noisy Labels
• CoDiM: Learning with Noisy Labels via Contrastive Semi-Supervised Learning