Observation-specific explanations through scattered data approximation
This work addresses the need for interpretable AI by providing observation-specific explanations, but it appears incremental as it builds on existing scattered data approximation techniques.
The paper tackles the problem of explaining black-box model predictions by assigning importance scores to individual data points, and it validates the proposed method on simulated and real-world datasets.
This work introduces the definition of observation-specific explanations to assign a score to each data point proportional to its importance in the definition of the prediction process. Such explanations involve the identification of the most influential observations for the black-box model of interest. The proposed method involves estimating these explanations by constructing a surrogate model through scattered data approximation utilizing the orthogonal matching pursuit algorithm. The proposed approach is validated on both simulated and real-world datasets.