Towards Aggregating Weighted Feature Attributions
This work addresses the need for more comprehensive model explanations in machine learning, though it appears incremental as it combines existing explanation classes.
The paper tackles the problem of explaining machine learning models by proposing AVA, an algorithm that fuses antecedent event influence and value attribution to provide both local and global feature attributions, with experimentation showing it convincingly favors this approach.
Current approaches for explaining machine learning models fall into two distinct classes: antecedent event influence and value attribution. The former leverages training instances to describe how much influence a training point exerts on a test point, while the latter attempts to attribute value to the features most pertinent to a given prediction. In this work, we discuss an algorithm, AVA: Aggregate Valuation of Antecedents, that fuses these two explanation classes to form a new approach to feature attribution that not only retrieves local explanations but also captures global patterns learned by a model. Our experimentation convincingly favors weighting and aggregating feature attributions via AVA.