Contrastive Explanations for Explaining Model Adaptations

TL;DR

This paper introduces a framework for explaining model adaptations over time using contrastive explanations, addressing the challenge of interpreting non-static AI systems and their changes.

Contribution

It proposes a novel framework and method for automatically identifying data regions affected by model updates, enhancing interpretability of evolving models.

Findings

Framework effectively explains model adaptations

Method accurately identifies affected data regions

Empirical evaluation demonstrates practical usefulness

Abstract

Many decision making systems deployed in the real world are not static - a phenomenon known as model adaptation takes place over time. The need for transparency and interpretability of AI-based decision models is widely accepted and thus have been worked on extensively. Usually, explanation methods assume a static system that has to be explained. Explaining non-static systems is still an open research question, which poses the challenge how to explain model adaptations. In this contribution, we propose and (empirically) evaluate a framework for explaining model adaptations by contrastive explanations. We also propose a method for automatically finding regions in data space that are affected by a given model adaptation and thus should be explained.