Conditional validity of inductive conformal predictors

Research output: Contribution to journalArticlepeer-review

Abstract

Conformal predictors are set predictors that are automatically valid in the sense of having coverage probability equal to or exceeding a given condence level. Inductive conformal predictors are a computationally efficient version of conformal predictors satisfying the same property of validity. However, inductive conformal predictors have been only known to control unconditional coverage probability. This paper explores various versions of conditional validity and various ways to achieve them using inductive conformal predictors and their modications.
Original languageEnglish
Pages (from-to)475-490
Number of pages16
JournalJMLR: Workshop and Conference Proceedings
Volume25
Publication statusPublished - 17 Nov 2012

Keywords

  • inductive conformal predictors
  • conditional validity
  • batch mode of learning
  • boosting
  • MART
  • spam detection

Cite this