We present a unified framework of provably efficient randomized estimators for Shapley values. We also provide, for the first time to our knowledge, provable guarantees for KernelSHAP, the most popular Shapley value method.

Using coding theory and heavy tails to design powerful watermarks with zero distortion.