MinShap: A Modified Shapley Value Approach for Feature Selection

TL;DR

MinShap is a novel feature selection method that modifies Shapley values by considering minimum marginal contributions, improving accuracy and stability over existing algorithms.

Contribution

It introduces MinShap, a new approach based on the minimum marginal contribution concept, with theoretical guarantees and enhanced performance in various settings.

Findings

MinShap outperforms LOCO, GCM, and Lasso in accuracy and stability.

Theoretical guarantees for Type I error and faithfulness in DAG models.

Algorithms using p-value perspective improve performance in low-sample scenarios.

Abstract

Feature selection is a classical problem in statistics and machine learning, and it continues to remain an extremely challenging problem especially in the context of unknown non-linear relationships with dependent features. On the other hand, Shapley values are a classic solution concept from cooperative game theory that is widely used for feature attribution in general non-linear models with highly-dependent features. However, Shapley values are not naturally suited for feature selection since they tend to capture both direct effects from each feature to the response and indirect effects through other features. In this paper, we combine the advantages of Shapley values and adapt them to feature selection by proposing \emph{MinShap}, a modification of the Shapley value framework along with a suite of other related algorithms. In particular for MinShap, instead of taking the average marginal contributions over permutations of features, considers the minimum marginal contribution across permutations. We provide a theoretical foundation motivated by the faithfulness assumption in DAG (directed acyclic graphical models), a guarantee for the Type I error of MinShap, and show through numerical simulations and real data experiments that MinShap tends to outperform state-of-the-art feature selection algorithms such as LOCO, GCM and Lasso in terms of both accuracy and stability. We also introduce a suite of algorithms related to MinShap by using the multiple testing/p-value perspective that improves performance in lower-sample settings and provide supporting theoretical guarantees.