AI RESEARCH
QuadraSHAP: Stable and Scalable Shapley Values for Product Games via Gauss-Legendre Quadrature
arXiv CS.LG
•
ArXi:2605.05870v1 Announce Type: new We study the efficient computation of Shapley values for \emph{product games} -- cooperative games in which the coalition value factorizes as a product of per-player terms. Such games arise in machine learning explainability whenever the value function inherits a multiplicative structure from the underlying model, as in kernel methods with product kernels and tree-based models.