mlsanityData to 7 Oct 2026

Minimax risk classifiers with 0-1 loss

arXiv 2201.06487 ↗First posted 17 Jan 2022stat.ML cs.LGcited by 9

Not on the map. Why?

Abstract (arXiv)

Supervised classification techniques use training samples to learn a classification rule with small expected 0-1 loss (error probability). Conventional methods enable tractable learning and provide out-of-sample generalization by using surrogate losses instead of the 0-1 loss and considering specific families of rules (hypothesis classes). This paper presents minimax risk classifiers (MRCs) that minize the worst-case 0-1 loss with respect to uncertainty sets of distributions that can include the underlying distribution, with a tunable confidence. We show that MRCs can provide tight performance guarantees at learning and are strongly universally consistent using feature mappings given by characteristic kernels. The paper also proposes efficient optimization techniques for MRC learning and shows that the methods presented can provide accurate classification together with tight performance guarantees in practice.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Builds on 24 of its 78 references that are in the map. First posted on arXiv 17 Jan 2022.

Builds on

24 of 78 references.

References · 24

  1. 2021Convex optimizationcited by 2,479
  2. 2016Kernel Mean Embedding of Distributions: A Review and Beyondcited by 360
  3. 2016Statistics of Robust Optimization: A Generalized Empirical Likelihood Approachcited by 182
  4. 2020Estimating means of bounded random variables by bettingcited by 164
  5. 2016Variance-based regularization with convex objectivescited by 153
  6. 2015On the Equivalence between Kernel Quadrature Rules and Random Feature Expansionscited by 136
  7. 2019Mean estimation and regression under heavy-tailed distributions--a surveycited by 133
  8. 2015Distributionally Robust Logistic Regressioncited by 124
  9. 2017Regularization via Mass Transportationcited by 79
  10. 2019Distributionally Robust Optimization and Generalization in Kernel Methodscited by 73
  11. 2013Loss minimization and parameter estimation with heavy tailscited by 50
  12. 2015Risk Bounds for the Majority Vote: From a PAC-Bayesian Analysis to a Learning Algorithmcited by 50
12 more
  1. 2019PAC-Bayes Un-Expected Bernstein Inequalitycited by 36
  2. 2016A Minimax Approach to Supervised Learningcited by 33
  3. 2021PAC-Bayes, MAC-Bayes and Conditional Mutual Information: Fast rate bounds that handle general VC classescited by 31
  4. 2019Incorporating Unlabeled Data into Distributionally Robust Learningcited by 20
  5. 2015RSG: Beating Subgradient Method without Smoothness and Strong Convexitycited by 15
  6. 2020Generalized Maximum Entropy for Supervised Classificationcited by 12
  7. 2020Minimax Classification with 0-1 Loss and Performance Guaranteescited by 11
  8. 2015Optimally Combining Classifiers Using Unlabeled Datacited by 9
  9. 2020The Strength of Nesterov's Extrapolation in the Individual Convergence of Nonsmooth Optimizationcited by 9
  10. 2015Optimal Binary Classifier Aggregation for General Lossescited by 6
  11. 2021MRCpy: A Library for Minimax Risk Classifierscited by 5
  12. 2019General Supervision via Probabilistic Transformationscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 3; 2024: 0; 2025: 1; 2026: 4.

9 papers.

  1. 2023Double-Weighting for Covariate Shift Adaptationcited by 7
  2. 2022Minimax Classification under Concept Drift with Multidimensional Adaptation and Performance Guaranteescited by 3
  3. 2023Efficient Learning of Minimax Risk Classifiers in High Dimensionscited by 2
  4. 2023Minimax Forward and Backward Learning of Evolving Tasks with Performance Guaranteescited by 2
  5. 2026Safe Fairness Guarantees Without Demographics in Classification: Spectral Uncertainty Set Perspectivecited by 1
  6. 2025Efficient Large-Scale Learning of Minimax Risk Classifierscited by 0
  7. 2026Minimax Generalized Cross-Entropycited by 0
  8. 2026NL-PAC: Specification Ambiguity and Certified Minimax Risk Floors in LLM-Mediated Supervisioncited by 0
  9. 2026IIns-VAE+: A Robust Transfer Learning Framework for Environmental Identification in Wireless Sensingcited by 0