Neighborhood Collective Estimation for Noisy Label Identification and Correction

TL;DR

This paper introduces Neighborhood Collective Estimation, a novel approach for identifying and correcting noisy labels in training data by leveraging local neighborhood information, significantly improving model robustness on benchmark datasets.

Contribution

The paper proposes a neighborhood-based method to re-estimate sample reliability and correct noisy labels, reducing confirmation bias in noisy label learning.

Findings

Outperforms state-of-the-art methods on CIFAR-10, CIFAR-100, Clothing-1M, and Webvision-1.0.

Effectively separates clean and noisy samples using neighborhood consensus.

Enhances model performance through improved label correction techniques.

Abstract

Learning with noisy labels (LNL) aims at designing strategies to improve model performance and generalization by mitigating the effects of model overfitting to noisy labels. The key success of LNL lies in identifying as many clean samples as possible from massive noisy data, while rectifying the wrongly assigned noisy labels. Recent advances employ the predicted label distributions of individual samples to perform noise verification and noisy label correction, easily giving rise to confirmation bias. To mitigate this issue, we propose Neighborhood Collective Estimation, in which the predictive reliability of a candidate sample is re-estimated by contrasting it against its feature-space nearest neighbors. Specifically, our method is divided into two steps: 1) Neighborhood Collective Noise Verification to separate all training samples into a clean or noisy subset, 2) Neighborhood Collective Label Correction to relabel noisy samples, and then auxiliary techniques are used to assist further model optimization. Extensive experiments on four commonly used benchmark datasets, i.e., CIFAR-10, CIFAR-100, Clothing-1M and Webvision-1.0, demonstrate that our proposed method considerably outperforms state-of-the-art methods.