Abstract:Learning against label noise is a vital topic to guarantee a reliable performance for deep neural networks.Recent research usually refers to dynamic noise modeling with model output probabilities and loss values, and then separates clean and noisy samples.These methods have gained notable success. However, unlike cherry-picked data, existing approaches often cannot perform well when facing imbalanced datasets, a common scenario in the real world.We thoroughly investigate this phenomenon and point out two major issues that hinder the performance, i.e., inter-class loss distribution discrepancy and misleading predictions due to uncertainty.The first issue is that existing methods often perform class-agnostic noise modeling. However, loss distributions show a significant discrepancy among classes under class imbalance, and class-agnostic noise modeling can easily get confused with noisy samples and samples in minority classes.The second issue refers to that models may output misleading predictions due to epistemic uncertainty and aleatoric uncertainty, thus existing methods that rely solely on the output probabilities may fail to distinguish confident samples. Inspired by our observations, we propose an Uncertainty-aware Label Correction framework(ULC) to handle label noise on imbalanced datasets. First, we perform epistemic uncertainty-aware class-specific noise modeling to identify trustworthy clean samples and refine/discard highly confident true/corrupted labels.Then, we introduce aleatoric uncertainty in the subsequent learning process to prevent noise accumulation in the label noise modeling process. We conduct experiments on several synthetic and real-world datasets. The results demonstrate the effectiveness of the proposed method, especially on imbalanced datasets.

Imbalanced Binary Classification under Distribution Uncertainty

Identifying Incorrect Classifications with Balanced Uncertainty

Striking the Right Balance with Uncertainty

Leveraging Uncertainty Estimates To Improve Classifier Performance

Binary Classification with Imbalanced Data

Learning Confidence Bounds for Classification with Imbalanced Data

Distribution-free uncertainty quantification for classification under label shift

Class Uncertainty: A Measure to Mitigate Class Imbalance

Uncertainty Voting Ensemble for Imbalanced Deep Regression

Imbalanced Data Classification Based on Hybrid Methods

A Conjugate Property between Loss Functions and Uncertainty Sets in Classification Problems

Binary Classification: Counterbalancing Class Imbalance by Applying Regression Models in Combination with One-Sided Label Shifts

Uncertainty-Aware Learning against Label Noise on Imbalanced Datasets

A Bayesian classifier for uncertain data.

A Structural Svm Based Approach for Binary Classification under Class Imbalance

GEV Regression with Convex Loss Applied to Imbalanced Binary Classification.

A composite entropy-based uncertainty measure guided attribute reduction for imbalanced mixed-type data

An Adaptive Pre-clustering Support Vector Machine for Binary Imbalanced Classification.

BSM loss: A superior way in modeling aleatory uncertainty of fine_grained classification

Support Vector Machine-Based Optimized Decision Threshold Adjustment Strategy for Classifying Imbalanced Data