How Much Can Stratification Improve the Approximation of Shapley Values?

Kolpaczki, Patrick; Haselbeck, Georg und Hüllermeier, Eyke (Juli 2024): How Much Can Stratification Improve the Approximation of Shapley Values? Explainable Artificial Intelligence, Valletta, Malta, 17. –19. July 2024. Longo, Luca; Lapuschkin, Sebastian und Seifert, Christin (Hrsg.): In: Communications in Computer and Information Science, Bd. 2154 Springer, Cham. S. 489-512

Abstract

Over the last decade, the Shapley value has become one of the most widely applied tools to provide post-hoc explanations for black box models. However, its theoretically justified solution to the problem of dividing a collective benefit to the members of a group, such as features or data points, comes at a price. Without strong assumptions, the exponential number of member subsets excludes an exact calculation of the Shapley value. In search for a remedy, recent works have demonstrated the efficacy of approximations based on sampling with stratification, in which the sample space is partitioned into smaller subpopulations. The effectiveness of this technique mainly depends on the degree to which the allocation of available samples over the formed strata mirrors their unknown variances. To uncover the hypothetical potential of stratification, we investigate the gap in approximation quality caused by the lack of knowledge of the optimal allocation. Moreover, we combine recent advances to propose two state-of-the-art algorithms Adaptive SVARM and Continuous Adaptive SVARM that adjust the sample allocation on-the-fly. The potential of our approach is assessed in an empirical evaluation.

• Dokumententyp: Konferenzbeitrag (Paper)
• Keywords: Shapley Value, Cooperative Games, Explainable Artificial Intelligence,Feature Importance
• Fakultät: Mathematik, Informatik und Statistik > Informatik > Künstliche Intelligenz und Maschinelles Lernen
• Themengebiete: 000 Informatik, Informationswissenschaft, allgemeine Werke > 004 Informatik
• ISBN: 978-3-031-63797-1