Efficient SAGE Estimation via Causal Structure Learning

by   Christoph Luther, et al.

The Shapley Additive Global Importance (SAGE) value is a theoretically appealing interpretability method that fairly attributes global importance to a model's features. However, its exact calculation requires the computation of the feature's surplus performance contributions over an exponential number of feature sets. This is computationally expensive, particularly because estimating the surplus contributions requires sampling from conditional distributions. Thus, SAGE approximation algorithms only take a fraction of the feature sets into account. We propose d-SAGE, a method that accelerates SAGE approximation. d-SAGE is motivated by the observation that conditional independencies (CIs) between a feature and the model target imply zero surplus contributions, such that their computation can be skipped. To identify CIs, we leverage causal structure learning (CSL) to infer a graph that encodes (conditional) independencies in the data as d-separations. This is computationally more efficient because the expense of the one-time graph inference and the d-separation queries is negligible compared to the expense of surplus contribution evaluations. Empirically we demonstrate that d-SAGE enables the efficient and accurate estimation of SAGE values.


page 1

page 2

page 3

page 4


Explanatory causal effects for model agnostic explanations

This paper studies the problem of estimating the contributions of featur...

Understanding Global Feature Contributions Through Additive Importance Measures

Understanding the inner workings of complex machine learning models is a...

Fast, Dense Feature SDM on an iPhone

In this paper, we present our method for enabling dense SDM to run at ov...

The Causal Topography of Cognition

The causal structure of cognition can be simulated but not implemented c...

Approximating the Shapley Value without Marginal Contributions

The Shapley value is arguably the most popular approach for assigning a ...

Computing SHAP Efficiently Using Model Structure Information

SHAP (SHapley Additive exPlanations) has become a popular method to attr...

WeightedSHAP: analyzing and improving Shapley based feature attributions

Shapley value is a popular approach for measuring the influence of indiv...

Please sign up or login with your details

Forgot password? Click here to reset