Explanatory Model Monitoring to Understand the Effects of Feature Shifts on Performance

TL;DR

This paper introduces XPE, a novel method combining Optimal Transport and Shapley Values to explain performance degradation in models due to feature shifts, providing actionable insights across multiple data modalities.

Contribution

The paper proposes a new explainability approach for model monitoring that attributes performance changes to input features, enhancing understanding of model deterioration under feature shifts.

Findings

XPE outperforms baselines on various datasets.

It provides interpretable insights into feature shifts.

Applicable across images, audio, and tabular data.

Abstract

Monitoring and maintaining machine learning models are among the most critical challenges in translating recent advances in the field into real-world applications. However, current monitoring methods lack the capability of provide actionable insights answering the question of why the performance of a particular model really degraded. In this work, we propose a novel approach to explain the behavior of a black-box model under feature shifts by attributing an estimated performance change to interpretable input characteristics. We refer to our method that combines concepts from Optimal Transport and Shapley Values as Explanatory Performance Estimation (XPE). We analyze the underlying assumptions and demonstrate the superiority of our approach over several baselines on different data sets across various data modalities such as images, audio, and tabular data. We also indicate how the generated results can lead to valuable insights, enabling explanatory model monitoring by revealing potential root causes for model deterioration and guiding toward actionable countermeasures.