Tractable Shapley Values and Interactions via Tensor Networks

TL;DR

This paper introduces TN-SHAP, a tensor network-based method that efficiently computes Shapley values and interactions with significantly fewer evaluations, providing theoretical guarantees and substantial speedups over existing methods.

Contribution

The paper presents a novel tensor network approach to efficiently approximate Shapley values and interactions, reducing computational complexity and providing theoretical error bounds.

Findings

Matches enumeration accuracy on UCI datasets

Achieves 25-1000x speedups over KernelSHAP-IQ

Reduces evaluation cost by orders of magnitude

Abstract

We show how to replace the O(2^n) coalition enumeration over n features behind Shapley values and Shapley-style interaction indices with a few-evaluation scheme on a tensor-network (TN) surrogate: TN-SHAP. The key idea is to represent a predictor's local behavior as a factorized multilinear map, so that coalitional quantities become linear probes of a coefficient tensor. TN-SHAP replaces exhaustive coalition sweeps with just a small number of targeted evaluations to extract order-k Shapley interactions. In particular, both order-1 (single-feature) and order-2 (pairwise) computations have cost O(n*poly(chi) + n^2), where chi is the TN's maximal cut rank. We provide theoretical guarantees on the approximation error and tractability of TN-SHAP. On UCI datasets, our method matches enumeration on the fitted surrogate while reducing evaluation by orders of magnitude and achieves 25-1000x wall-clock speedups over KernelSHAP-IQ at comparable accuracy, while amortizing training across local cohorts.