Best Transition Matrix Esitimation or Best Label Noise Robustness Classifier? Two Possible Methods to Enhance the Performance of T-revision

2025-01-02

AI Summary

To address performance degradation caused by label noise, this paper focuses on accurate estimation of the noise transition matrix and co-design of robust classifiers. We propose two novel variants—T-Revision-Alpha and T-Revision-Softmax—that enhance the stability of the original T-Revision method under both known and unknown noise structures. We introduce the first joint paradigm of “transition matrix estimation–classifier optimization”, integrating anchor-point-assumption-driven matrix estimation, forward correction, and importance reweighting. Experiments demonstrate significant improvements in transition matrix estimation accuracy and classification accuracy on FashionMNIST under known noise. On CIFAR-10 with unknown real-world noise, our method achieves state-of-the-art clean-label prediction performance, validating its generalizability and robustness in practical noisy-label scenarios.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Intelligent Robots: State Estimation
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Label noise refers to incorrect labels in a dataset caused by human errors or collection defects, which is common in real-world applications and can significantly reduce the accuracy of models. This report explores how to estimate noise transition matrices and construct deep learning classifiers that are robust against label noise. In cases where the transition matrix is known, we apply forward correction and importance reweighting methods to correct the impact of label noise using the transition matrix. When the transition matrix is unknown or inaccurate, we use the anchor point assumption and T-Revision series methods to estimate or correct the noise matrix. In this study, we further improved the T-Revision method by developing T-Revision-Alpha and T-Revision-Softmax to enhance stability and robustness. Additionally, we designed and implemented two baseline classifiers, a Multi-Layer Perceptron (MLP) and ResNet-18, based on the cross-entropy loss function. We compared the performance of these methods on predicting clean labels and estimating transition matrices using the FashionMINIST dataset with known noise transition matrices. For the CIFAR-10 dataset, where the noise transition matrix is unknown, we estimated the noise matrix and evaluated the ability of the methods to predict clean labels.

Problem

• Label Noise
• Deep Learning Accuracy
• Noise Transition Matrix Estimation

Innovation

• T-Correction Method
• MLP and ResNet-18 Classifiers
• Noise Robustness