English

Computing Full Conformal Prediction Set with Approximate Homotopy

Machine Learning 2019-11-11 v2 Machine Learning Optimization and Control

Abstract

If you are predicting the label yy of a new object with y^\hat y, how confident are you that y=y^y = \hat y? Conformal prediction methods provide an elegant framework for answering such question by building a 100(1α)%100 (1 - \alpha)\% confidence region without assumptions on the distribution of the data. It is based on a refitting procedure that parses all the possibilities for yy to select the most likely ones. Although providing strong coverage guarantees, conformal set is impractical to compute exactly for many regression problems. We propose efficient algorithms to compute conformal prediction set using approximated solution of (convex) regularized empirical risk minimization. Our approaches rely on a new homotopy continuation technique for tracking the solution path with respect to sequential changes of the observations. We also provide a detailed analysis quantifying its complexity.

Keywords

Cite

@article{arxiv.1909.09365,
  title  = {Computing Full Conformal Prediction Set with Approximate Homotopy},
  author = {Eugene Ndiaye and Ichiro Takeuchi},
  journal= {arXiv preprint arXiv:1909.09365},
  year   = {2019}
}

Comments

Conference on Neural Information Processing Systems (NeurIPS), 2019

R2 v1 2026-06-23T11:21:04.218Z