Abstract
Conformal predictors are set predictors that are automatically valid in the sense of having coverage probability equal to or exceeding a given confidence level. Inductive conformal predictors are a computationally efficient version of conformal predictors satisfying the same property of validity. However, inductive conformal predictors have only been 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 modifications. In particular, it discusses a convenient expression of one of the modifications in terms of ROC curves.
Original language | English |
---|---|
Pages (from-to) | 349 - 376 |
Number of pages | 28 |
Journal | Machine Learning |
Volume | 92 |
Issue number | 2-3 |
Early online date | 8 May 2013 |
DOIs | |
Publication status | Published - 5 Jul 2013 |
Keywords
- Inductive conformal predictors
- Conditional validity
- Batch mode of learning
- ROC curves
- Boosting
- MART
- Spam detection