mlsanityData to 7 Oct 2026

Deep AUC Maximization for Medical Image Classification: Challenges and Opportunities

arXiv 2111.02400 ↗First posted 1 Nov 2021cs.LG cs.AI cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

In this extended abstract, we will present and discuss opportunities and challenges brought about by a new deep learning method by AUC maximization (aka underlinebf Deep underlinebf AUC underlinebf Maximization or bf DAM) for medical image classification. Since AUC (aka area under ROC curve) is a standard performance measure for medical image classification, hence directly optimizing AUC could achieve a better performance for learning a deep neural network than minimizing a traditional loss function (e.g., cross-entropy loss). Recently, there emerges a trend of using deep AUC maximization for large-scale medical image classification. In this paper, we will discuss these recent results by highlighting (i) the advancements brought by stochastic non-convex optimization algorithms for DAM; (ii) the promising results on various medical image classification problems. Then, we will discuss challenges and opportunities of DAM for medical image classification from three perspectives, feature learning, large-scale optimization, and learning trustworthy AI models.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 31 of its 63 references that are in the map. First posted on arXiv 1 Nov 2021.

Builds on

31 of 63 references.

References · 31

  1. 2019A Survey on Bias and Fairness in Machine Learningcited by 1,911
  2. 2018This Looks Like That: Deep Learning for Interpretable Image Recognitioncited by 731
  3. 2020Contrastive Learning of Medical Visual Representations from Paired Images and Textcited by 636
  4. 2019Nuanced Metrics for Measuring Unintended Bias with Real Data for Text Classificationcited by 396
  5. 2020Fairness in Machine Learning: A Surveycited by 369
  6. 2021Big Self-Supervised Models Advance Medical Image Classificationcited by 320
  7. 2018AI Fairness 360: An Extensible Toolkit for Detecting, Understanding, and Mitigating Unwanted Algorithmic Biascited by 319
  8. 2019On Gradient Descent Ascent for Nonconvex-Concave Minimax Problemscited by 266
  9. 2020CheXclusion: Fairness gaps in deep chest X-ray classifierscited by 153
  10. 2020Smooth-AP: Smoothing the Path Towards Large-Scale Image Retrievalcited by 98
  11. 2020Stochastic Recursive Gradient Descent Ascent for Stochastic Nonconvex-Strongly-Concave Minimax Problemscited by 75
  12. 2019Interpretable Image Recognition with Hierarchical Prototypescited by 64
19 more
  1. 2019Stochastic AUC Maximization with Deep Neural Networkscited by 61
  2. 2020MoCo-CXR: MoCo Pretraining Improves Representation and Transferability of Chest X-ray Modelscited by 61
  3. 2019Optimizing Rank-based Metrics with Blackbox Differentiationcited by 58
  4. 2013One-Pass AUC Optimizationcited by 42
  5. 2019The Fairness of Risk Scores Beyond Classification: Bipartite Ranking and the xAUC Metriccited by 41
  6. 2016Scalable Learning of Non-Decomposable Objectivescited by 40
  7. 2019Towards Accurate One-Stage Object Detection with AP-Losscited by 36
  8. 2016End-to-end training of object class detectors for mean average precisioncited by 35
  9. 2021Using StyleGAN for Visual Interpretability of Deep Learning Models on Medical Imagescited by 35
  10. 2020Communication-Efficient Distributed Stochastic AUC Maximization with Deep Neural Networkscited by 27
  11. 2020Alternating proximal-gradient steps for (stochastic) nonconvex-concave minimax problemscited by 26
  12. 2020A Ranking-based, Balanced Loss Function Unifying Classification and Localisation in Object Detectioncited by 25
  13. 2020AP-Loss for Accurate One-Stage Object Detectioncited by 24
  14. 2013On the Generalization Ability of Online Learning Algorithms for Pairwise Loss Functionscited by 23
  15. 2014Online and Stochastic Gradient Methods for Non-decomposable Loss Functionscited by 21
  16. 2021Federated Deep AUC Maximization for Heterogeneous Data with a Constant Communication Complexitycited by 21
  17. 2021Technical Challenges for Training Fair Neural Networkscited by 20
  18. 2020Fast Objective & Duality Gap Convergence for Non-Convex Strongly-Concave Min-Max Problems with PL Conditioncited by 19
  19. 2016Efficient Optimization for Rank-based Loss Functionscited by 15

Built on by

Cited by year

Citing papers per year: 2022: 2; 2023: 0; 2024: 1; 2025: 3.

6 papers.

  1. 2024A Closer Look at AUROC and AUPRC under Class Imbalancecited by 56
  2. 2022NEVIS'22: A Stream of 100 Tasks Sampled from 30 Years of Computer Vision Researchcited by 18
  3. 2025Preserving AUC Fairness in Learning with Noisy Protected Groupscited by 6
  4. 2022Asymptotically Unbiased Instance-wise Regularized Partial AUC Optimization: Theory and Algorithmcited by 5
  5. 2025Simulating the Unseen: Crash Prediction Must Learn from What Did Not Happencited by 5
  6. 2025Closing the Approximation Gap of Partial AUC Optimization: A Tale of Two Formulationscited by 0