{"date_created":"2025-01-16T16:12:16Z","title":"KernelSHAP-IQ: Weighted Least Square Optimization for Shapley Interactions","publisher":"PMLR","page":"14308–14342","series_title":"Proceedings of Machine Learning Research","intvolume":"       235","publication":"Proceedings of the 41st International Conference on Machine Learning","user_id":"93420","author":[{"last_name":"Fumagalli","full_name":"Fumagalli, Fabian","first_name":"Fabian"},{"first_name":"Maximilian","full_name":"Muschalik, Maximilian","last_name":"Muschalik"},{"first_name":"Patrick","last_name":"Kolpaczki","full_name":"Kolpaczki, Patrick"},{"first_name":"Eyke","last_name":"Hüllermeier","full_name":"Hüllermeier, Eyke"},{"last_name":"Hammer","full_name":"Hammer, Barbara","first_name":"Barbara"}],"type":"conference","volume":235,"year":"2024","citation":{"ieee":"F. Fumagalli, M. Muschalik, P. Kolpaczki, E. Hüllermeier, and B. Hammer, “KernelSHAP-IQ: Weighted Least Square Optimization for Shapley Interactions,” in <i>Proceedings of the 41st International Conference on Machine Learning</i>, 2024, vol. 235, pp. 14308–14342.","ama":"Fumagalli F, Muschalik M, Kolpaczki P, Hüllermeier E, Hammer B. KernelSHAP-IQ: Weighted Least Square Optimization for Shapley Interactions. In: Salakhutdinov R, Kolter Z, Heller K, et al., eds. <i>Proceedings of the 41st International Conference on Machine Learning</i>. Vol 235. Proceedings of Machine Learning Research. PMLR; 2024:14308–14342.","bibtex":"@inproceedings{Fumagalli_Muschalik_Kolpaczki_Hüllermeier_Hammer_2024, series={Proceedings of Machine Learning Research}, title={KernelSHAP-IQ: Weighted Least Square Optimization for Shapley Interactions}, volume={235}, booktitle={Proceedings of the 41st International Conference on Machine Learning}, publisher={PMLR}, author={Fumagalli, Fabian and Muschalik, Maximilian and Kolpaczki, Patrick and Hüllermeier, Eyke and Hammer, Barbara}, editor={Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, year={2024}, pages={14308–14342}, collection={Proceedings of Machine Learning Research} }","mla":"Fumagalli, Fabian, et al. “KernelSHAP-IQ: Weighted Least Square Optimization for Shapley Interactions.” <i>Proceedings of the 41st International Conference on Machine Learning</i>, edited by Ruslan Salakhutdinov et al., vol. 235, PMLR, 2024, pp. 14308–14342.","chicago":"Fumagalli, Fabian, Maximilian Muschalik, Patrick Kolpaczki, Eyke Hüllermeier, and Barbara Hammer. “KernelSHAP-IQ: Weighted Least Square Optimization for Shapley Interactions.” In <i>Proceedings of the 41st International Conference on Machine Learning</i>, edited by Ruslan Salakhutdinov, Zico Kolter, Katherine Heller, Adrian Weller, Nuria Oliver, Jonathan Scarlett, and Felix Berkenkamp, 235:14308–14342. Proceedings of Machine Learning Research. PMLR, 2024.","short":"F. Fumagalli, M. Muschalik, P. Kolpaczki, E. Hüllermeier, B. Hammer, in: R. Salakhutdinov, Z. Kolter, K. Heller, A. Weller, N. Oliver, J. Scarlett, F. Berkenkamp (Eds.), Proceedings of the 41st International Conference on Machine Learning, PMLR, 2024, pp. 14308–14342.","apa":"Fumagalli, F., Muschalik, M., Kolpaczki, P., Hüllermeier, E., &#38; Hammer, B. (2024). KernelSHAP-IQ: Weighted Least Square Optimization for Shapley Interactions. In R. Salakhutdinov, Z. Kolter, K. Heller, A. Weller, N. Oliver, J. Scarlett, &#38; F. Berkenkamp (Eds.), <i>Proceedings of the 41st International Conference on Machine Learning</i> (Vol. 235, pp. 14308–14342). PMLR."},"department":[{"_id":"660"}],"status":"public","date_updated":"2025-01-16T16:19:03Z","_id":"58223","language":[{"iso":"eng"}],"project":[{"_id":"109","name":"TRR 318: TRR 318 - Erklärbarkeit konstruieren","grant_number":"438445824"},{"name":"TRR 318 - C: TRR 318 - Project Area C","_id":"117"},{"name":"TRR 318 - C3: TRR 318 - Subproject C3","_id":"126"}],"editor":[{"full_name":"Salakhutdinov, Ruslan","last_name":"Salakhutdinov","first_name":"Ruslan"},{"last_name":"Kolter","full_name":"Kolter, Zico","first_name":"Zico"},{"first_name":"Katherine","last_name":"Heller","full_name":"Heller, Katherine"},{"first_name":"Adrian","full_name":"Weller, Adrian","last_name":"Weller"},{"first_name":"Nuria","last_name":"Oliver","full_name":"Oliver, Nuria"},{"first_name":"Jonathan","full_name":"Scarlett, Jonathan","last_name":"Scarlett"},{"first_name":"Felix","full_name":"Berkenkamp, Felix","last_name":"Berkenkamp"}],"abstract":[{"lang":"eng","text":"The Shapley value (SV) is a prevalent approach of allocating credit to machine learning (ML) entities to understand black box ML models. Enriching such interpretations with higher-order interactions is inevitable for complex systems, where the Shapley Interaction Index (SII) is a direct axiomatic extension of the SV. While it is well-known that the SV yields an optimal approximation of any game via a weighted least square (WLS) objective, an extension of this result to SII has been a long-standing open problem, which even led to the proposal of an alternative index. In this work, we characterize higher-order SII as a solution to a WLS problem, which constructs an optimal approximation via SII and k-Shapley values (k-SII). We prove this representation for the SV and pairwise SII and give empirically validated conjectures for higher orders. As a result, we propose KernelSHAP-IQ, a direct extension of KernelSHAP for SII, and demonstrate state-of-the-art performance for feature interactions."}]}