Collaborative Label Correction via Entropy Thresholding

TL;DR

This paper introduces a collaborative label correction method using entropy thresholding, which leverages low-entropy predictions for more reliable supervision, improving training with noisy labels in deep neural networks.

Contribution

The paper proposes a novel entropy-based collaborative framework that enhances label correction and outperforms existing methods on various benchmarks.

Findings

Low-entropy predictions are more reliable than noisy labels.

The proposed CLC method outperforms state-of-the-art approaches.

Maintains more training samples than previous methods.

Abstract

Deep neural networks (DNNs) have the capacity to fit extremely noisy labels nonetheless they tend to learn data with clean labels first and then memorize those with noisy labels. We examine this behavior in light of the Shannon entropy of the predictions and demonstrate the low entropy predictions determined by a given threshold are much more reliable as the supervision than the original noisy labels. It also shows the advantage in maintaining more training samples than previous methods. Then, we power this entropy criterion with the Collaborative Label Correction (CLC) framework to further avoid undesired local minimums of the single network. A range of experiments have been conducted on multiple benchmarks with both synthetic and real-world settings. Extensive results indicate that our CLC outperforms several state-of-the-art methods.