Conformal Risk Control for Non-Monotonic Losses
Not on the map. Why?
Abstract (arXiv)
Conformal risk control is an extension of conformal prediction for controlling risk functions beyond miscoverage. The original algorithm controls the expected value of a loss that is monotonic in a one-dimensional parameter. Here, we present risk control guarantees for generic algorithms applied to possibly non-monotonic losses with multidimensional parameters. The guarantees depend on the stability of the algorithm -- unstable algorithms have looser guarantees. We give applications of this technique to selective image classification, FDR and IOU control of tumor segmentations, and multigroup debiasing of recidivism predictions across overlapping race and sex groups using empirical risk minimization.
Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 25 of its 49 references that are in the map. First posted on arXiv 23 Feb 2026.
Builds on
25 of 49 references.
References · 25
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2021A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantificationcited by 777
- 2016Distribution-Free Predictive Inference For Regressioncited by 657
- 2015Train faster, generalize better: Stability of stochastic gradient descentcited by 547
- 2019Conformal Prediction Under Covariate Shiftcited by 511
- 2020PraNet: Parallel Reverse Attention Network for Polyp Segmentationcited by 460
- 2022Conformal prediction beyond exchangeabilitycited by 348
- 2022Conformal Risk Controlcited by 290
- 2021Learn then Test: Calibrating Predictive Algorithms to Achieve Risk Controlcited by 256
- 2018Multiaccuracy: Black-Box Post-Processing for Fairness in Classificationcited by 221
- 2023Conformal Prediction With Conditional Guaranteescited by 146
- 2013Learning by Transductioncited by 120
13 more
- 2024Theoretical Foundations of Conformal Predictioncited by 96
- 2019Sharper bounds for uniformly stable algorithmscited by 91
- 2013A Conformal Prediction Approach to Explore Functional Datacited by 85
- 2022Batch Multivalid Conformal Predictioncited by 66
- 2022Practical Adversarial Multivalid Conformal Predictioncited by 65
- 2018Generalization Bounds for Uniformly Stable Algorithmscited by 56
- 2023How to Trust Your Diffusion Model: A Convex Optimization Approach to Conformal Risk Controlcited by 50
- 2024Conformal Validity Guarantees Exist for Any Data Distribution (and How to Find Them)cited by 26
- 2022Efficiently Controlling Multiple Risks with Pareto Testingcited by 21
- 2024Automatically Adaptive Conformal Risk Controlcited by 16
- 2021Black-box tests for algorithmic stabilitycited by 15
- 2025Gradient Equilibrium in Online Learning: Theory and Applicationscited by 12
- 2020Post-Selection Inference via Algorithmic Stabilitycited by 7
Built on by
Citing papers per year: 2026: 5.
5 papers.
- 2026Conformal Policy Controlcited by 5
- 2026A Joint Finite-Sample Certificate for Adaptive Selective Conformal Risk Controlcited by 3
- 2026Conformal Risk Control under Non-Monotone Losses: Theory and Finite-Sample Guaranteescited by 2
- 2026Risk-Controlled Post-Processing of Decision Policiescited by 1
- 2026Certified Domain Consistency for Multi-Domain Retrieval: Label-Free Per-Domain Contamination Control with Conformal Risk Guaranteescited by 0
- 2026Certified Domain Consistency for Multi-Domain Retrieval: Label-Free Per-Domain Contamination Control with Conformal Risk Guarantees14 Jul 2026
- 2026A Joint Finite-Sample Certificate for Adaptive Selective Conformal Risk Control7 Jun 2026
- 2026Risk-Controlled Post-Processing of Decision Policies7 May 2026
- 2026Conformal Risk Control under Non-Monotone Losses: Theory and Finite-Sample Guarantees2 Apr 2026
- 2026Conformal Policy Control2 Mar 2026