AUC Optimization

AUC optimization focuses on directly maximizing the area under the ROC curve, a crucial performance metric for classifiers, particularly in imbalanced datasets. Current research emphasizes robust AUC optimization under weak supervision, including scenarios with noisy labels, incomplete data, and multiple unlabeled datasets, employing techniques like partial AUC and adversarial training to address these challenges. These advancements improve classifier performance and robustness, impacting various applications from recommendation systems to medical diagnosis where accurate ranking of instances is paramount. Furthermore, efficient algorithms are being developed to scale AUC optimization to large datasets, addressing the computational cost of pairwise comparisons.

Papers

May 25, 2023

AUC Optimization from Multiple Unlabeled Datasets
Zheng Xie, Yu Liu, Ming Li
Weak Supervision Unitary Gradient Neural Network AUC Optimization

May 23, 2023

Weakly Supervised AUC Optimization: A Unified Partial AUC Approach
Zheng Xie, Yu Liu, Hao-Yuan He, Ming Li, Zhi-Hua Zhou
Weak Supervision Unified Alignment Partial AUC AUC Optimization

May 2, 2023

Early Classifying Multimodal Sequences
Alexander Cao, Jean Utke, Diego Klabjan
Classification Code Unimodal Model Multimodal Classification Early Classification Multimodal Sequence AUC Optimization

April 17, 2023

Enhancing Personalized Ranking With Differentiable Group AUC Optimization
Xiao Sun, Bo Zhang, Chenrui Zhang, Han Ren, Mingchen Cai
Recommendation Model Interest Loss Personalized Ranking AUC Metric AUC Optimization

June 24, 2022

AdAUC: End-to-end Adversarial AUC Optimization Against Long-tail Problems
Wenzheng Hou, Qianqian Xu, Zhiyong Yang, Shilong Bao, Yuan He, Qingming Huang
Adversarial Example Adversarial Training Long Tailed Problem AUC Optimization Regularization Scheme

January 4, 2022

Evolutionary Multitasking AUC Optimization
Chao Wang, Kai Wu, Jing Liu
Pairwise Learning AUC Optimization