[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117309-en":3,"doc-seo-117309-105":29,"detail-sidebar-cat-0-en-105":81},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":20,"language":21,"language_code":22,"site_id":23,"html_lang":22,"table_of_contents":24,"faqs":25,"seo_title":26,"seo_description":14,"update_tm":27,"read_time":28},117309,687207024643,"Oliver","https://ap-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2",6,"Technology","shapiq：Shapley Interactions for Machine Learning - Explaining models with Shapley Interactions","Explains how to measure feature interactions for black-box machine learning models using Shapley interaction concepts beyond single-feature attributions. Introduces shapiq as an easy-to-use framework supporting agnostic and tree-based explainers, enabling interpretation from Shapley values (order 1) up to k-order interactions and Möbius interactions. Covers implementation targets across frameworks, provides faithfulness and complexity considerations, and summarizes benchmark design with 11 ML domains plus 100 benchmark games for evaluating approximators and exact computers.","shapiq: Shapley Interactions for Machine Learning  \nMaximilian Muschalik 1 , Hubert Baniecki 2 , Fabian Fumagalli 3 , Patrick Kolpaczki 4 , Barbara Hammer 3 , and Eyke Hüllermeier 1  \nHow do I measure interactions between multiple features for black box models beyond feature attributions?  \nI want to use Shapley values for other ML applications. How do I compute them?  \nExplain Models with Shapley Interactions Game Theory for General ML Applications  \nExplaining models with shapiq is easy:  \n➢ Agnostic Explainerand Imputers  \n➢ Tree Explainer  \n“Does the location of my property affect its price?” “Why is this a dog?”  \n“How does my language model predict a positive sentiment?”  \nAny Model  \n(e.g. , torch, sklearn,…)  \ns  \nTree Model  \n(e.g. , xgboost lightgbm,…)  \nInterpretation: Shapley interactions generalize the Shapley value beyond individual effects up to any-order and capture synergies between features.  \norder 1: Shapley Value  \nup to order k: Shapley Interactions  \nFaithfulness and Complexity  \nup to order n: Möbius Interactions  \nAny Value Function  \n(as a callable)  \nshapiq includes:  \n➢ 20 concepts (Shapley value and interactions, Banzhaf value and interactions, Faithful Shapley, Generalized values, Möbius, Core,…)  \n➢ 14 state-of-the-art approximators and exact computers  \n\n| Class | Shapley Interactions Shapley Values |\n| --- | --- |\n| Approximator | KernelSHAP-IQ KernelSHAP |\n|  | Inconsistent KernelSHAP-IQ kADD-SHAP |\n|  | Faith-SHAP Owen Sampling |\n|  | SHAP-IQ Unbiased KernelSHAP |\n|  | SVARM-IQ SVARM |\n|  | Permutation Sampling (SII) Permutation Sampling (SV) |\n|  | Permutation Sampling (STII) Stratified Sampling |\n| Computer | Möbius Converter\u003Cbr>Exact Computer |\n\nBenchmark of 11 ML domains (e.g. , explanation, data valuation, uncertainty quantification,…)  \nGames: 100 benchmark games with more than 2000 pre-computed configurations  \nEvaluation of Approximators on the Benchmark  \n➢  \n➢  \nperformance depends on the domain kernel and stratification based on top  \n1) LMU Munich  \n2)  \nUniversity of Warsaw and  \nWarsaw University of Technology  \n3) Bielefeld University 4) Paderborn University  \nmmschlk/shapiq  \nPR Welcome!","cbCaisQmo7pESQww","https://ap.wps.com/l/cbCaisQmo7pESQww","pdf",1716163,1,"English","en",105,"# Explaining models with Shapley Interactions\n## From Shapley values to higher-order interactions\n## shapiq explainers and model compatibility\n## Benchmark design and evaluation","[{\"question\":\"How are approximation methods evaluated in this presentation?\",\"answer\":\"Approximators are benchmarked across 11 ML domains using 100 benchmark games with 2000+ pre-computed configurations. Evaluation results consider performance dependencies on the domain kernel and stratification scheme.\"}]","shapiq：Shapley Interactions for Machine Learning - Explaining models with Shapley Interactions | PDF",1785675107,3,{"code":4,"msg":30,"data":31},"ok",{"site_id":23,"language":22,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":76,"head_meta":78,"extra_data":80,"updated_unix":27},"shapiq-shapley-interactions-for-machine-learning-explaining-models-with-shapley-interactions","",{"@graph":35,"@context":75},[36,52,66],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,49],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":28},"https://docshare.wps.com/document/technology/",{"item":50,"name":13,"@type":42,"position":51},"https://docshare.wps.com/document/shapiq-shapley-interactions-for-machine-learning-explaining-models-with-shapley-interactions/117309/",4,{"url":50,"name":13,"@type":53,"author":54,"headline":13,"publisher":56,"fileFormat":59,"inLanguage":22,"description":14,"dateModified":60,"datePublished":60,"encodingFormat":59,"isAccessibleForFree":61,"interactionStatistic":62},"DigitalDocument",{"name":9,"@type":55},"Person",{"url":40,"name":57,"@type":58},"DocShare","Organization","application/pdf","2026-08-02",true,{"@type":63,"interactionType":64,"userInteractionCount":4},"InteractionCounter",{"@type":65},"ViewAction",{"@type":67,"mainEntity":68},"FAQPage",[69],{"name":70,"@type":71,"acceptedAnswer":72},"How are approximation methods evaluated in this presentation?","Question",{"text":73,"@type":74},"Approximators are benchmarked across 11 ML domains using 100 benchmark games with 2000+ pre-computed configurations. Evaluation results consider performance dependencies on the domain kernel and stratification scheme.","Answer","https://schema.org",{"og:url":50,"og:type":77,"og:title":13,"og:site_name":57,"og:description":14},"article",{"robots":79,"canonical":50},"index,follow",{"doc_id":7,"site_id":23},{"code":4,"msg":5,"data":82},[83,87,91,95,100,103,108,113,118,121,125],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":84,"show_sort_weight":85,"slug":86},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":88,"show_sort_weight":89,"slug":90},"Literature",80,"literature",{"id":51,"doc_module":4,"doc_module_name":45,"category_name":92,"show_sort_weight":93,"slug":94},"Exam",70,"exam",{"id":96,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},5,"Comic",60,"comic",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":101,"slug":102},50,"technology",{"id":104,"doc_module":4,"doc_module_name":45,"category_name":105,"show_sort_weight":106,"slug":107},7,"Healthcare",40,"healthcare",{"id":109,"doc_module":4,"doc_module_name":45,"category_name":110,"show_sort_weight":111,"slug":112},8,"Research & Report",30,"research-report",{"id":114,"doc_module":4,"doc_module_name":45,"category_name":115,"show_sort_weight":116,"slug":117},9,"Religion & Spirituality",20,"religion-spirituality",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":119,"show_sort_weight":116,"slug":120},"World Cup","world-cup",{"id":122,"doc_module":4,"doc_module_name":45,"category_name":123,"show_sort_weight":122,"slug":124},10,"Lifestyle","lifestyle",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":96,"slug":128},19,"General","general"]