mlsanityData to 14 Oct 2026

Extreme Conformal Prediction: Reliable Intervals for High-Impact Events

arXiv 2505.08578 ↗First posted 13 May 2025stat.ME stat.AP stat.MLcited by 5

Not on the map. Why?

Abstract (arXiv)

Conformal prediction is a popular method to construct prediction intervals with marginal coverage guarantees from black-box machine learning models. In applications with potentially high-impact events, such as flooding or financial crises, regulators often require very high confidence for such intervals. However, if the desired level of confidence is too large relative to the amount of data used for calibration, then classical conformal methods provide infinitely wide, thus, uninformative prediction intervals. In this paper, we propose a new method to overcome this limitation. We bridge extreme value statistics and conformal prediction to provide reliable and informative prediction intervals with high-confidence coverage, which can be constructed using any black-box extreme quantile regression method. A weighted version of our approach can account for nonstationary data. The advantages of our extreme conformal prediction method are illustrated in a simulation study and in an application to flood risk forecasting.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 25 of its 68 references that are in the map. First posted on arXiv 13 May 2025.

Builds on

25 of 68 references.

References · 25

  1. 2016Distribution-Free Predictive Inference For Regressioncited by 668
  2. 2019Conformalized Quantile Regressioncited by 629
  3. 2019Conformal Prediction Under Covariate Shiftcited by 519
  4. 2022Conformal prediction beyond exchangeabilitycited by 355
  5. 2019Predictive inference with the jackknife+cited by 211
  6. 2018High-Quality Prediction Intervals for Deep Learning: A Distribution-Free, Ensembled Approachcited by 117
  7. 2019Nested conformal prediction and quantile out-of-bag ensemble methodscited by 89
  8. 2014Regression Conformal Prediction with Nearest Neighbourscited by 59
  9. 2020Predictive Inference Is Free with the Jackknife+-after-Bootstrapcited by 55
  10. 2020Discriminative Jackknife: Quantifying Uncertainty in Deep Learning via Higher-Order Influence Functionscited by 34
  11. 2024Conformal Prediction: A Data Perspectivecited by 25
  12. 2022Neural Networks for Extreme Quantile Regression with an Application to Forecasting of Flood Riskcited by 19
13 more
  1. 2021Gradient boosting for extreme quantile regressioncited by 17
  2. 2021Learning Prediction Intervals for Regression: Generalization and Calibrationcited by 16
  3. 2020Sparse Structures for Multivariate Extremescited by 12
  4. 2022Regression modelling of spatiotemporal extreme U.S. wildfires via partially-interpretable neural networkscited by 11
  5. 2015Extremes on river networkscited by 9
  6. 2022Extremal Random Forestscited by 9
  7. 2023On Regression in Extreme Regionscited by 6
  8. 2024Progression: an extrapolation principle for regressioncited by 6
  9. 2018Improving precipitation forecasts using extreme quantile regressioncited by 4
  10. 2020An Extreme Value Bayesian Lasso for the Conditional Left and Right Tailscited by 4
  11. 2021Causal Modelling of Heavy-Tailed Variables and Confounders with Application to River Flowcited by 4
  12. 2018A horse racing between the block maxima method and the peak-over-threshold approachcited by 3
  13. 2022Modeling panels of extremescited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 5.

5 papers.

  1. 2026Extrapolation in Statistical Learning with Extreme Value Theorycited by 1
  2. 2026On Optimal Data Splitting for Split Conformal Predictioncited by 1
  3. 2026Geometry of tail allocation in conformal prediction intervalscited by 0
  4. 2026Stop Suppressing the Tail: Causal Inference for Extreme Eventscited by 0
  5. 2026Reaching the Tail: Calibration Diversity Drives Conformal Coverage under Data Scarcitycited by 0