CD-split and HPD-split: efficient conformal regions in high dimensions

Abstract

Conformal methods create prediction bands that control average coverage assuming solely i.i.d. data. Although the literature has mostly focused on prediction intervals, more general regions can often better represent uncertainty. For instance, a bimodal target is better represented by the union of two intervals. Such prediction regions are obtained by CD-split , which combines the split method and a data-driven partition of the feature space which scales to high dimensions. CD-split however contains many tuning parameters, and their role is not clear. In this paper, we provide new insights on CD-split by exploring its theoretical properties. In particular, we show that CD-split converges asymptotically to the oracle highest predictive density set and satisfies local and asymptotic conditional validity. We also present simulations that show how to tune CD-split. Finally, we introduce HPD-split, a variation of CD-split that requires less tuning, and show that it shares the same theoretical guarantees as CD-split. In a wide variety of our simulations, CD-split and HPD-split have better conditional coverage and yield smaller prediction regions than other methods.

arXiv comment34 pages, 15 figures

References in corpus (11)

• TPZ : Photometric redshift PDFs and ancillary information by using prediction trees and random forests
• Least Ambiguous Set-Valued Classifiers with Bounded Error Levels
• Evaluation of probabilistic photometric redshift estimation approaches for The Rubin Observatory Legacy Survey of Space and Time (LSST)
• On the realistic validation of photometric redshifts, or why Teddy will never be Happy
• Nonparametric Conditional Density Estimation in a High-Dimensional Regression Setting
• Conformalized Quantile Regression
• What Lies Beneath: Using p(z) to Reduce Systematic Photometric Redshift Errors
• Conditional Density Estimation Tools in Python and R with Applications to Photometric Redshifts and Likelihood-Free Cosmological Inference
• Gaussian Process Conditional Density Estimation
• A Unified Framework for Constructing, Tuning and Assessing Photometric Redshift Density Estimates in a Selection Bias Setting
• Conformal prediction with localization

Cited by in corpus (5)

• Improving Conditional Coverage via Orthogonal Quantile Regression
• Calibrated Multiple-Output Quantile Regression with Representation Learning
• MD-split+: Practical Local Conformal Inference in High Dimensions
• Conformal inference for random objects
• Neural interval-censored survival regression with feature selection