Efficient Shapley Performance Attribution for Least-Squares Regression

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Bell, Logan, Devanathan, Nikhil, Boyd, Stephen
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929493562621952
author Bell, Logan
Devanathan, Nikhil
Boyd, Stephen
author_facet Bell, Logan
Devanathan, Nikhil
Boyd, Stephen
contents We consider the performance of a least-squares regression model, as judged by out-of-sample $R^2$. Shapley values give a fair attribution of the performance of a model to its input features, taking into account interdependencies between features. Evaluating the Shapley values exactly requires solving a number of regression problems that is exponential in the number of features, so a Monte Carlo-type approximation is typically used. We focus on the special case of least-squares regression models, where several tricks can be used to compute and evaluate regression models efficiently. These tricks give a substantial speed up, allowing many more Monte Carlo samples to be evaluated, achieving better accuracy. We refer to our method as least-squares Shapley performance attribution (LS-SPA), and describe our open-source implementation.
format Preprint
id arxiv_https___arxiv_org_abs_2310_19245
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Efficient Shapley Performance Attribution for Least-Squares Regression
Bell, Logan
Devanathan, Nikhil
Boyd, Stephen
Computation
62-08 (Primary), 62-04, 62J99 (Secondary)
We consider the performance of a least-squares regression model, as judged by out-of-sample $R^2$. Shapley values give a fair attribution of the performance of a model to its input features, taking into account interdependencies between features. Evaluating the Shapley values exactly requires solving a number of regression problems that is exponential in the number of features, so a Monte Carlo-type approximation is typically used. We focus on the special case of least-squares regression models, where several tricks can be used to compute and evaluate regression models efficiently. These tricks give a substantial speed up, allowing many more Monte Carlo samples to be evaluated, achieving better accuracy. We refer to our method as least-squares Shapley performance attribution (LS-SPA), and describe our open-source implementation.
title Efficient Shapley Performance Attribution for Least-Squares Regression
topic Computation
62-08 (Primary), 62-04, 62J99 (Secondary)
url https://arxiv.org/abs/2310.19245