Jo-SNC: Combating Noisy Labels Through Fostering Self- and Neighbor-Consistency.

2025-12-22

IEEE Transactions on Pattern Analysis and Machine Intelligence

AI Summary

This work addresses the performance degradation in supervised deep learning caused by label noise, encompassing both in-distribution and out-of-distribution noise. To tackle this challenge, the authors propose Jo-SNC, a method that integrates sample selection with model regularization. Jo-SNC identifies clean samples by evaluating prediction consistency between each sample and its nearest neighbors, and employs partial label learning and negative learning strategies to handle different noise types. The approach innovatively introduces an adaptive threshold mechanism based on Jensen–Shannon divergence and designs a triplet consistency regularization to enhance both prediction and feature stability. Extensive experiments on multiple benchmark datasets demonstrate that Jo-SNC significantly outperforms current state-of-the-art methods, confirming its effectiveness and robustness.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Search and Optimization: Learning to Search
• Natural Language Processing: Learning & Optimization for NLP

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Label noise is pervasive in various real-world scenarios, posing challenges in supervised deep learning. Deep networks are vulnerable to such label-corrupted samples due to the memorization effect. One major stream of previous methods concentrates on identifying clean data for training. However, these methods often neglect imbalances in label noise across different mini-batches and devote insufficient attention to out-of-distribution noisy data. To this end, we propose a noise-robust method named Jo-SNC (Joint sample selection and model regularization based on Self- and Neighbor-Consistency). Specifically, we propose to employ the Jensen-Shannon divergence to measure the "likelihood" of a sample being clean or out-of-distribution. This process factors in the nearest neighbors of each sample to reinforce the reliability of clean sample identification. We design a self-adaptive, data-driven thresholding scheme to adjust per-class selection thresholds. While clean samples undergo conventional training, detected in-distribution and out-of-distribution noisy samples are trained following partial label learning and negative learning, respectively. Finally, we advance the model performance further by proposing a triplet consistency regularization that promotes self-prediction consistency, neighbor-prediction consistency, and feature consistency. Extensive experiments on various benchmark datasets and comprehensive ablation studies demonstrate the effectiveness and superiority of our approach over existing state-of-the-art methods. Our code and models have been made publicly available at https://github.com/NUST-Machine-Intelligence-Laboratory/Jo-SNC.

Problem

• label noise
• noisy labels
• supervised deep learning
• out-of-distribution
• class imbalance

Innovation

• label noise
• self-consistency
• neighbor-consistency
• Jensen-Shannon divergence
• triplet consistency regularization