XAudit : A Theoretical Look at Auditing with Explanations
This work addresses the problem of auditing models for undesirable properties, which is crucial for responsible AI, but it is incremental as it builds on existing explanation methods.
The paper formalizes the role of explanations in auditing machine learning models and investigates their effectiveness, finding that counterfactual explanations are extremely helpful for auditing linear classifiers and decision trees for feature sensitivity, with concrete results showing benefits in average-case scenarios.
Responsible use of machine learning requires models to be audited for undesirable properties. While a body of work has proposed using explanations for auditing, how to do so and why has remained relatively ill-understood. This work formalizes the role of explanations in auditing and investigates if and how model explanations can help audits. Specifically, we propose explanation-based algorithms for auditing linear classifiers and decision trees for feature sensitivity. Our results illustrate that Counterfactual explanations are extremely helpful for auditing. While Anchors and decision paths may not be as beneficial in the worst-case, in the average-case they do aid a lot.