REPID: Regional Effect Plots with implicit Interaction Detection

Julia Herbinger, Bernd Bischl, Giuseppe Casalicchio
Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, PMLR 151:10209-10233, 2022.

Abstract

Machine learning models can automatically learn complex relationships, such as non-linear and interaction effects. Interpretable machine learning methods such as partial dependence plots visualize marginal feature effects but may lead to misleading interpretations when feature interactions are present. Hence, employing additional methods that can detect and measure the strength of interactions is paramount to better understand the inner workings of machine learning models. We demonstrate several drawbacks of existing global interaction detection approaches, characterize them theoretically, and evaluate them empirically. Furthermore, we introduce regional effect plots with implicit interaction detection, a novel framework to detect interactions between a feature of interest and other features. The framework also quantifies the strength of interactions and provides interpretable and distinct regions in which feature effects can be interpreted more reliably, as they are less confounded by interactions. We prove the theoretical eligibility of our method and show its applicability on various simulation and real-world examples.

Cite this Paper

BibTeX
@InProceedings{pmlr-v151-herbinger22a, title = { REPID: Regional Effect Plots with implicit Interaction Detection }, author = {Herbinger, Julia and Bischl, Bernd and Casalicchio, Giuseppe}, booktitle = {Proceedings of The 25th International Conference on Artificial Intelligence and Statistics}, pages = {10209--10233}, year = {2022}, editor = {Camps-Valls, Gustau and Ruiz, Francisco J. R. and Valera, Isabel}, volume = {151}, series = {Proceedings of Machine Learning Research}, month = {28--30 Mar}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v151/herbinger22a/herbinger22a.pdf}, url = {https://proceedings.mlr.press/v151/herbinger22a.html}, abstract = { Machine learning models can automatically learn complex relationships, such as non-linear and interaction effects. Interpretable machine learning methods such as partial dependence plots visualize marginal feature effects but may lead to misleading interpretations when feature interactions are present. Hence, employing additional methods that can detect and measure the strength of interactions is paramount to better understand the inner workings of machine learning models. We demonstrate several drawbacks of existing global interaction detection approaches, characterize them theoretically, and evaluate them empirically. Furthermore, we introduce regional effect plots with implicit interaction detection, a novel framework to detect interactions between a feature of interest and other features. The framework also quantifies the strength of interactions and provides interpretable and distinct regions in which feature effects can be interpreted more reliably, as they are less confounded by interactions. We prove the theoretical eligibility of our method and show its applicability on various simulation and real-world examples. } }
Endnote
%0 Conference Paper %T REPID: Regional Effect Plots with implicit Interaction Detection %A Julia Herbinger %A Bernd Bischl %A Giuseppe Casalicchio %B Proceedings of The 25th International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2022 %E Gustau Camps-Valls %E Francisco J. R. Ruiz %E Isabel Valera %F pmlr-v151-herbinger22a %I PMLR %P 10209--10233 %U https://proceedings.mlr.press/v151/herbinger22a.html %V 151 %X Machine learning models can automatically learn complex relationships, such as non-linear and interaction effects. Interpretable machine learning methods such as partial dependence plots visualize marginal feature effects but may lead to misleading interpretations when feature interactions are present. Hence, employing additional methods that can detect and measure the strength of interactions is paramount to better understand the inner workings of machine learning models. We demonstrate several drawbacks of existing global interaction detection approaches, characterize them theoretically, and evaluate them empirically. Furthermore, we introduce regional effect plots with implicit interaction detection, a novel framework to detect interactions between a feature of interest and other features. The framework also quantifies the strength of interactions and provides interpretable and distinct regions in which feature effects can be interpreted more reliably, as they are less confounded by interactions. We prove the theoretical eligibility of our method and show its applicability on various simulation and real-world examples.
APA
Herbinger, J., Bischl, B. & Casalicchio, G.. (2022). REPID: Regional Effect Plots with implicit Interaction Detection . Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 151:10209-10233 Available from https://proceedings.mlr.press/v151/herbinger22a.html.

Related Material