PUB - Publikationen an der Universität Bielefeld

Addressing the limitations of individual at- tribution scores via the Shapley value (SV), the field of explainable AI (XAI) has recently explored intricate interactions of features or data points. In particular, extensions of the SV, such as the Shapley Interaction Index (SII), have been proposed as a measure to still benefit from the axiomatic basis of the SV. However, similar to the SV, their exact computation remains computationally pro- hibitive. Hence, we propose with SVARM-IQ a sampling-based approach to efficiently ap- proximate Shapley-based interaction indices of any order. SVARM-IQ can be applied to a broad class of interaction indices, includ- ing the SII, by leveraging a novel stratified representation. We provide non-asymptotic theoretical guarantees on its approximation quality and empirically demonstrate that SVARM-IQ achieves state-of-the-art estima- tion results in practical XAI scenarios on dif- ferent model classes and application domains.