mlsanityData to 7 Oct 2026

Set-valued classification -- overview via a unified framework

arXiv 2102.12318 ↗First posted 24 Feb 2021stat.ML cs.LGcited by 27

Not on the map. Why?

Abstract (arXiv)

Multi-class classification problem is among the most popular and well-studied statistical frameworks. Modern multi-class datasets can be extremely ambiguous and single-output predictions fail to deliver satisfactory performance. By allowing predictors to predict a set of label candidates, set-valued classification offers a natural way to deal with this ambiguity. Several formulations of set-valued classification are available in the literature and each of them leads to different prediction strategies. The present survey aims to review popular formulations using a unified statistical framework. The proposed framework encompasses previously considered and leads to new formulations as well as it allows to understand underlying trade-offs of each formulation. We provide infinite sample optimal set-valued classification strategies and review a general plug-in principle to construct data-driven algorithms. The exposition is supported by examples and pointers to both theoretical and practical contributions. Finally, we provide experiments on real-world datasets comparing these approaches in practice and providing general practical guidelines.

Key facts. Cited by 27 papers in the map (data through 7 Oct 2026). Builds on 14 of its 60 references that are in the map. First posted on arXiv 24 Feb 2021.

Builds on

14 of 60 references.

References · 14

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,441
  3. 2017On Calibration of Modern Neural Networkscited by 4,860
  4. 2016Aggregated Residual Transformations for Deep Neural Networkscited by 2,540
  5. 2016Distribution-Free Predictive Inference For Regressioncited by 657
  6. 2020Classification with Valid and Adaptive Coveragecited by 390
  7. 2016Least Ambiguous Set-Valued Classifiers with Bounded Error Levelscited by 270
  8. 2019Presence-Only Geographical Priors for Fine-Grained Image Classificationcited by 109
  9. 2018Smooth Loss Functions for Deep Top-k Classificationcited by 46
  10. 2019On the Calibration of Multiclass Classification with Rejectioncited by 41
  11. 2015Loss Functions for Top-k Error: Analysis and Insightscited by 38
  12. 2019On the Consistency of Top-k Surrogate Lossescited by 35
2 more
  1. 2015Top-k Multiclass SVMcited by 28
  2. 2017On Reject and Refine Options in Multicategory Classificationcited by 9

Built on by

Cited by year

Citing papers per year: 2021: 3; 2022: 6; 2023: 4; 2024: 10; 2025: 4.

27 papers.

Sort:

By area:

  1. 2021Multi-Label Learning from Single Positive Labelscited by 81
  2. 2022Improving Expert Predictions with Conformal Predictioncited by 54
  3. 2023Predictor-Rejector Multi-Class Abstention: Theoretical Analysis and Algorithmscited by 36
  4. 2024Conformal Prediction Sets Improve Human Decision Makingcited by 35
  5. 2022Rank-based Decomposable Losses in Machine Learning: A Surveycited by 26
  6. 2023Designing Decision Support Systems Using Counterfactual Prediction Setscited by 26
  7. 2024Towards Human-AI Complementarity with Prediction Setscited by 25
  8. 2022Differentiable Top-k Classification Learningcited by 21
  9. 2022Improving Screening Processes via Calibrated Subset Selectioncited by 16
  10. 2024Conformal Prediction Sets Can Cause Disparate Impactcited by 14
  11. 2022Stochastic smoothing of the top-K calibrated hinge loss for deep imbalanced classificationcited by 10
  12. 2024Controlling Counterfactual Harm in Decision Support Systems Based on Prediction Setscited by 9
15 more
  1. 2024Building a stable classifier with the inflated argmaxcited by 8
  2. 2024Adapting Prediction Sets to Distribution Shifts Without Labelscited by 6
  3. 2024On the Impact of Uncertainty and Calibration on Likelihood-Ratio Membership Inference Attackscited by 4
  4. 2021Posthoc Verification and the Fallibility of the Ground Truthcited by 3
  5. 2022Set-valued prediction in hierarchical classification with constrained representation complexitycited by 3
  6. 2023Earning Extra Performance from Restrictive Feedbackscited by 3
  7. 2025Conformal Set-based Human-AI Complementarity with Multiple Expertscited by 2
  8. 2024Conformal Predictions for Probabilistically Robust Scalable Machine Learning Classificationcited by 1
  9. 2025Assumption-free stability for ranking problemscited by 1
  10. 2021Taming Overconfident Prediction on Unlabeled Data from Hindsightcited by 0
  11. 2023A two-head loss function for deep Average-K classificationcited by 0
  12. 2024AI-based Mapping of the Conservation Status of Orchid Assemblages at Global Scalecited by 0
  13. 2024Conjunction Subspaces Test for Conformal and Selective Classificationcited by 0
  14. 2025Class conditional conformal prediction for multiple inputs by p-value aggregationcited by 0
  15. 2025Set to Be Fair: Demographic Parity Constraints for Set-Valued Classificationcited by 0