Glad to share that our paper “ShaplEIG: Bayesian Experimental Design for Shapley Value Estimation”, joint work with Fabian Fumagalli, Maximilian Muschalik, Bernd Bischl and @matthiasfeurer.bsky.social, has been accepted to ICML 2026!
Paper: arxiv.org/abs/2606.02247
arxiv.org
ShaplEIG: Bayesian Experimental Design for Shapley Value Estimation
Shapley values are a principled attribution measure widely used in interpretable machine learning, but their exact computation scales exponentially with the number of players, motivating a wide range ...