Shapley-value based inductive conformal prediction

William Lopez Jaramillo, Evgueni Smirnov
Proceedings of the Tenth Symposium on Conformal and Probabilistic Prediction and Applications, PMLR 152:52-71, 2021.

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 ϵ0.1 compared with those produced by standard conformity scores (i.e. similarity between true and predicted output values).

Cite this Paper


BibTeX
@InProceedings{pmlr-v152-jaramillo21a, title = {Shapley-value based inductive conformal prediction}, author = {Jaramillo, William Lopez and Smirnov, Evgueni}, booktitle = {Proceedings of the Tenth Symposium on Conformal and Probabilistic Prediction and Applications}, pages = {52--71}, year = {2021}, editor = {Carlsson, Lars and Luo, Zhiyuan and Cherubin, Giovanni and An Nguyen, Khuong}, volume = {152}, series = {Proceedings of Machine Learning Research}, month = {08--10 Sep}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v152/jaramillo21a/jaramillo21a.pdf}, url = {https://proceedings.mlr.press/v152/jaramillo21a.html}, 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 $\epsilon \leq 0.1$ compared with those produced by standard conformity scores (i.e. similarity between true and predicted output values).} }
Endnote
%0 Conference Paper %T Shapley-value based inductive conformal prediction %A William Lopez Jaramillo %A Evgueni Smirnov %B Proceedings of the Tenth Symposium on Conformal and Probabilistic Prediction and Applications %C Proceedings of Machine Learning Research %D 2021 %E Lars Carlsson %E Zhiyuan Luo %E Giovanni Cherubin %E Khuong An Nguyen %F pmlr-v152-jaramillo21a %I PMLR %P 52--71 %U https://proceedings.mlr.press/v152/jaramillo21a.html %V 152 %X 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 $\epsilon \leq 0.1$ compared with those produced by standard conformity scores (i.e. similarity between true and predicted output values).
APA
Jaramillo, W.L. & Smirnov, E.. (2021). Shapley-value based inductive conformal prediction. Proceedings of the Tenth Symposium on Conformal and Probabilistic Prediction and Applications, in Proceedings of Machine Learning Research 152:52-71 Available from https://proceedings.mlr.press/v152/jaramillo21a.html.

Related Material