Partial AUC

Partial Area Under the ROC Curve (AUC) focuses on optimizing a specific region of the ROC curve, offering a more nuanced evaluation of classifier performance than the full AUC, particularly when considering trade-offs between true and false positive rates. Current research emphasizes efficient algorithms for partial AUC optimization, including novel loss functions and gradient-based methods tailored for deep learning models, addressing challenges like weak supervision and computational scalability. This refined metric is proving valuable in diverse applications, from recommendation systems and anomaly detection to mitigating toxicity in language models, improving the effectiveness and robustness of machine learning across various domains.

Papers

October 20, 2024

A Novel Characterization of the Population Area Under the Risk Coverage Curve (AURC) and Rates of Finite Sample Estimators
Han Zhou, Jordy Van Landeghem, Teodora Popordanoska, Matthew B. Blaschko
Confidence Score Finite Sample Na\"ive Population Selective Classification Partial AUC

July 2, 2024

Whispering Experts: Neural Interventions for Toxicity Mitigation in Language Models
Xavier Suau, Pieter Delobelle, Katherine Metcalf, Armand Joulin, Nicholas Apostoloff, Luca Zappella, Pau Rodríguez
Language Model State of the Art Whisper Neuron Level Model Behavior Toxic Language Partial AUC Toxicity Mitigation

February 29, 2024

Lower-Left Partial AUC: An Effective and Efficient Optimization Metric for Recommendation
Wentao Shi, Chenxu Wang, Fuli Feng, Yang Zhang, Wenjie Wang, Junkang Wu, Xiangnan He
Effective Recommendation Recommendation System Recommendation Task AUC Metric Partial AUC

October 5, 2023

Non-Smooth Weakly-Convex Finite-sum Coupled Compositional Optimization
Quanqi Hu, Dixian Zhu, Tianbao Yang
Compositional Optimization Non Smooth Partial AUC

September 26, 2023

SeMAnD: Self-Supervised Anomaly Detection in Multimodal Geospatial Datasets
Daria Reshetova, Swetava Ganguli, C. V. Krishnakumar Iyer, Vipul Pandey
Anomaly Detection Self Supervised Anomaly Partial AUC Multimodal Geospatial

September 22, 2023

Document Understanding for Healthcare Referrals
Jimit Mistry, Natalia M. Arzeno
Open Domain Document Understanding Exam Document Partial AUC

July 26, 2023

EasyNet: An Easy Network for 3D Industrial Anomaly Detection
Ruitao Chen, Guoyang Xie, Jiaqi Liu, Jinbao Wang, Ziqi Luo, Jinfan Wang, Feng Zheng
Anomaly Detection Computer Vision Task Anomaly Map 3D Anomaly Unknown Network Partial AUC

June 26, 2023

Beyond AUROC & co. for evaluating out-of-distribution detection performance
Galadrielle Humblot-Renaux, Sergio Escalera, Thomas B. Moeslund
Distribution Detection Distribution Sample Understanding Out of Distribution Partial AUC

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

March 23, 2023

Failure-tolerant Distributed Learning for Anomaly Detection in Wireless Networks
Marc Katzef, Andrew C. Cullen, Tansu Alpcan, Christopher Leckie, Justin Kopacz
Native Robustness Anomaly Detection Wireless Network Distributed Learning Partial AUC

October 26, 2022

FeDXL: Provable Federated Learning for Deep X-Risk Optimization
Zhishuai Guo, Rong Jin, Jiebo Luo, Tianbao Yang
Empirical Risk Minimization Pairwise Loss Partial AUC

October 8, 2022

Asymptotically Unbiased Instance-wise Regularized Partial AUC Optimization: Theory and Algorithm
Huiyang Shao, Qianqian Xu, Zhiyong Yang, Shilong Bao, Qingming Huang
Practical Algorithm Theoretical Understanding Area Under the ROC Curve ROC Curve Partial AUC

June 23, 2022

Optimizing Two-way Partial AUC with an End-to-end Framework
Zhiyong Yang, Qianqian Xu, Shilong Bao, Yuan He, Xiaochun Cao, Qingming Huang
End to End Gradient Based Optimization ROC Curve Partial AUC Surrogate Optimization

March 1, 2022

When AUC meets DRO: Optimizing Partial AUC for Deep Learning with Non-Convex Convergence Guarantee
Dixian Zhu, Gang Li, Bokun Wang, Xiaodong Wu, Tianbao Yang
Deep Learning Robust Optimization Distributionally Robust Optimization Value at Risk Partial AUC AUC Metric Efficient Gradient