Confident Feature Ranking
This addresses the need for more reliable feature interpretation in ML models for stakeholders, though it is incremental as it builds on existing post-hoc importance methods.
The paper tackles the problem of unstable feature importance rankings in machine learning by introducing a framework to quantify uncertainty in global importance values, resulting in simultaneous confidence intervals for feature ranks that include the true ranks with high probability and enable selection of top-k features.
Machine learning models are widely applied in various fields. Stakeholders often use post-hoc feature importance methods to better understand the input features' contribution to the models' predictions. The interpretation of the importance values provided by these methods is frequently based on the relative order of the features (their ranking) rather than the importance values themselves. Since the order may be unstable, we present a framework for quantifying the uncertainty in global importance values. We propose a novel method for the post-hoc interpretation of feature importance values that is based on the framework and pairwise comparisons of the feature importance values. This method produces simultaneous confidence intervals for the features' ranks, which include the ``true'' (infinite sample) ranks with high probability, and enables the selection of the set of the top-k important features.