Randomized multi-class classification under system constraints: a unified approach via post-processing
Not on the map. Why?
Abstract (arXiv)
We study the problem of multi-class classification under system-level constraints expressible as linear functionals over randomized classifiers. We propose a post-processing approach that adjusts a given base classifier to satisfy general constraints without retraining. Our method formulates the problem as a linearly constrained stochastic program over randomized classifiers, and leverages entropic regularization and dual optimization techniques to construct a feasible solution. We provide finite-sample guarantees for the risk and constraint satisfaction for the final output of our algorithm under minimal assumptions. The framework accommodates a broad class of constraints, including fairness, abstention, and churn requirements.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 20 of its 40 references that are in the map. First posted on arXiv 16 Dec 2025.
Builds on
20 of 40 references.
References · 20
- 2021Convex optimizationcited by 2,479
- 2016Distribution-Free Predictive Inference For Regressioncited by 657
- 2018A Reductions Approach to Fair Classificationcited by 585
- 2016Least Ambiguous Set-Valued Classifiers with Bounded Error Levelscited by 270
- 2019Fair Regression: Quantitative Definitions and Reduction-based Algorithmscited by 168
- 2018Classification with Fairness Constraints: A Meta-Algorithm with Provable Guaranteescited by 146
- 2017On the Properties of the Softmax Function with Application in Game Theory and Reinforcement Learningcited by 136
- 2018Optimization with Non-Differentiable Constraints with Applications to Fairness, Recall, Churn, and Other Goalscited by 103
- 2019Pairwise Fairness for Ranking and Regressioncited by 58
- 2018How To Make the Gradients Small Stochastically: Even Faster Convex and Nonconvex SGDcited by 55
- 2020Classification with Rejection Based on Cost-sensitive Classificationcited by 55
- 2022Fair and Optimal Classification via Post-Processingcited by 50
8 more
- 2019The Complexity of Making the Gradient Small in Stochastic Convex Optimizationcited by 33
- 2021Classification with abstention but without disparitiescited by 27
- 2022Fair learning with Wasserstein barycenters for non-decomposable performance measurescited by 24
- 2019General Fair Empirical Risk Minimizationcited by 23
- 2022Fair Bayes-Optimal Classifiers Under Predictive Paritycited by 16
- 2022Consistent Multiclass Algorithms for Complex Metrics and Constraintscited by 13
- 2024Regression under demographic parity constraints via unlabeled post-processingcited by 5
- 2024Finite-Sample and Distribution-Free Fair Classification: Optimal Trade-off Between Excess Risk and Fairness, and the Cost of Group-Blindnesscited by 2
Built on by
Citing papers per year: 2026: 1.
1 paper.