Shapley Based Residual Decomposition for Instance Analysis

Tommy Liu, Amanda S Barnard
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:21375-21387, 2023.

Abstract

In this paper, we introduce the idea of decomposing the residuals of regression with respect to the data instances instead of features. This allows us to determine the effects of each individual instance on the model and each other, and in doing so makes for a model-agnostic method of identifying instances of interest. In doing so, we can also determine the appropriateness of the model and data in the wider context of a given study. The paper focuses on the possible applications that such a framework brings to the relatively unexplored field of instance analysis in the context of Explainable AI tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-liu23b, title = {Shapley Based Residual Decomposition for Instance Analysis}, author = {Liu, Tommy and Barnard, Amanda S}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {21375--21387}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/liu23b/liu23b.pdf}, url = {https://proceedings.mlr.press/v202/liu23b.html}, abstract = {In this paper, we introduce the idea of decomposing the residuals of regression with respect to the data instances instead of features. This allows us to determine the effects of each individual instance on the model and each other, and in doing so makes for a model-agnostic method of identifying instances of interest. In doing so, we can also determine the appropriateness of the model and data in the wider context of a given study. The paper focuses on the possible applications that such a framework brings to the relatively unexplored field of instance analysis in the context of Explainable AI tasks.} }
Endnote
%0 Conference Paper %T Shapley Based Residual Decomposition for Instance Analysis %A Tommy Liu %A Amanda S Barnard %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-liu23b %I PMLR %P 21375--21387 %U https://proceedings.mlr.press/v202/liu23b.html %V 202 %X In this paper, we introduce the idea of decomposing the residuals of regression with respect to the data instances instead of features. This allows us to determine the effects of each individual instance on the model and each other, and in doing so makes for a model-agnostic method of identifying instances of interest. In doing so, we can also determine the appropriateness of the model and data in the wider context of a given study. The paper focuses on the possible applications that such a framework brings to the relatively unexplored field of instance analysis in the context of Explainable AI tasks.
APA
Liu, T. & Barnard, A.S.. (2023). Shapley Based Residual Decomposition for Instance Analysis. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:21375-21387 Available from https://proceedings.mlr.press/v202/liu23b.html.

Related Material