PWSHAP: A Path-Wise Explanation Model for Targeted Variables

TL;DR

PWSHAP is a novel explanation framework that assesses the targeted effect of specific variables in complex models, especially useful in sensitive domains, by combining causal graphs with on-manifold Shapley values for transparent, robust analysis.

Contribution

It introduces Path-Wise Shapley effects (PWSHAP), integrating causal DAGs with Shapley values to target specific variables' effects in complex models, with theoretical error bounds and practical applications.

Findings

Effective local bias and mediation analysis.

Robustness to adversarial attacks.

Clear interpretability and true locality demonstrated.

Abstract

Predictive black-box models can exhibit high accuracy but their opaque nature hinders their uptake in safety-critical deployment environments. Explanation methods (XAI) can provide confidence for decision-making through increased transparency. However, existing XAI methods are not tailored towards models in sensitive domains where one predictor is of special interest, such as a treatment effect in a clinical model, or ethnicity in policy models. We introduce Path-Wise Shapley effects (PWSHAP), a framework for assessing the targeted effect of a binary (e.g.~treatment) variable from a complex outcome model. Our approach augments the predictive model with a user-defined directed acyclic graph (DAG). The method then uses the graph alongside on-manifold Shapley values to identify effects along causal pathways whilst maintaining robustness to adversarial attacks. We establish error bounds for the identified path-wise Shapley effects and for Shapley values. We show PWSHAP can perform local bias and mediation analyses with faithfulness to the model. Further, if the targeted variable is randomised we can quantify local effect modification. We demonstrate the resolution, interpretability, and true locality of our approach on examples and a real-world experiment.