Guarantee Regions for Local Explanations
This addresses reliability issues in interpretability for machine learning users, though it is incremental as it builds on existing local explanation methods.
The paper tackles the problem that local surrogate explanation methods like LIME lack guarantees for extrapolation to surrounding regions, and proposes an anchor-based algorithm to identify regions where explanations are guaranteed correct, demonstrating larger guarantee regions and detection of misleading explanations.
Interpretability methods that utilise local surrogate models (e.g. LIME) are very good at describing the behaviour of the predictive model at a point of interest, but they are not guaranteed to extrapolate to the local region surrounding the point. However, overfitting to the local curvature of the predictive model and malicious tampering can significantly limit extrapolation. We propose an anchor-based algorithm for identifying regions in which local explanations are guaranteed to be correct by explicitly describing those intervals along which the input features can be trusted. Our method produces an interpretable feature-aligned box where the prediction of the local surrogate model is guaranteed to match the predictive model. We demonstrate that our algorithm can be used to find explanations with larger guarantee regions that better cover the data manifold compared to existing baselines. We also show how our method can identify misleading local explanations with significantly poorer guarantee regions.