Shapley-value based inductive conformal prediction

William Lopez Jaramillo, Evgueni Smirnov

Research output: Chapter in Book/Report/Conference proceedingConference article in proceedingAcademicpeer-review

Abstract

Shapley values of individual instances were recently proposed for the problem of data valuation. They were defined as the average marginal instance contributions to the performance of a given predictor. In this paper we propose to use Shapley values of individual instances as conformity scores. To compute these values efficiently and exactly we employ a standard algorithm based on nearest neighbor classification and propose a variant of this algorithm for clustered data. Both variants are used for computing Shapley conformity scores for inductive conformal predictors. The experiments show that the Shapley-value conformity scores result in smaller prediction sets for significance level compared with those produced by standard conformity scores (i.e. similarity between true and predicted output values).
Original languageEnglish
Title of host publicationProceedings of Machine Learning Research
Subtitle of host publicationVolume 152: Conformal and Probabilistic Prediction and Applications
Pages52-71
Volume152
Publication statusPublished - 2021
Event10th Conformal and Probabilistic Prediction and Applications 2021 - Online, Centre for Reliable Machine Learning, Egham, United Kingdom
Duration: 8 Sep 202110 Sep 2021
https://cml.rhul.ac.uk/copa2021/

Conference

Conference10th Conformal and Probabilistic Prediction and Applications 2021
Abbreviated titleCOPA 2021
Country/TerritoryUnited Kingdom
CityEgham
Period8/09/2110/09/21
Internet address

Cite this