Learning with Noisy Labels Using Collaborative Sample Selection and Contrastive Semi-Supervised Learning

TL;DR

This paper introduces a novel collaborative sample selection method using CLIP and contrastive semi-supervised learning to effectively identify clean samples and reduce noise impact in learning with noisy labels, improving generalization.

Contribution

The paper proposes a new approach combining CLIP, Gaussian Mixture Models, and contrastive co-training to enhance clean sample selection and mitigate confirmation bias in noisy label learning.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Effectively removes noisy samples from the clean set.

Improves DNN generalization in noisy label scenarios.

Abstract

Learning with noisy labels (LNL) has been extensively studied, with existing approaches typically following a framework that alternates between clean sample selection and semi-supervised learning (SSL). However, this approach has a limitation: the clean set selected by the Deep Neural Network (DNN) classifier, trained through self-training, inevitably contains noisy samples. This mixture of clean and noisy samples leads to misguidance in DNN training during SSL, resulting in impaired generalization performance due to confirmation bias caused by error accumulation in sample selection. To address this issue, we propose a method called Collaborative Sample Selection (CSS), which leverages the large-scale pre-trained model CLIP. CSS aims to remove the mixed noisy samples from the identified clean set. We achieve this by training a 2-Dimensional Gaussian Mixture Model (2D-GMM) that combines the probabilities from CLIP with the predictions from the DNN classifier. To further enhance the adaptation of CLIP to LNL, we introduce a co-training mechanism with a contrastive loss in semi-supervised learning. This allows us to jointly train the prompt of CLIP and the DNN classifier, resulting in improved feature representation, boosted classification performance of DNNs, and reciprocal benefits to our Collaborative Sample Selection. By incorporating auxiliary information from CLIP and utilizing prompt fine-tuning, we effectively eliminate noisy samples from the clean set and mitigate confirmation bias during training. Experimental results on multiple benchmark datasets demonstrate the effectiveness of our proposed method in comparison with the state-of-the-art approaches.