Explaining Concept Drift through the Evolution of Group Counterfactuals
This addresses the problem of interpretability in dynamic environments for practitioners needing to diagnose model degradation, though it is incremental as it builds on existing drift detection and explanation techniques.
The paper tackles the challenge of explaining concept drift in machine learning models by introducing a method that analyzes the temporal evolution of group-based counterfactual explanations to reveal structural changes in decision boundaries and underlying rationale, enabling distinction between root causes like spatial data shifts or re-labeling.
Machine learning models in dynamic environments often suffer from concept drift, where changes in the data distribution degrade performance. While detecting this drift is a well-studied topic, explaining how and why the model's decision-making logic changes still remains a significant challenge. In this paper, we introduce a novel methodology to explain concept drift by analyzing the temporal evolution of group-based counterfactual explanations (GCEs). Our approach tracks shifts in the GCEs' cluster centroids and their associated counterfactual action vectors before and after a drift. These evolving GCEs act as an interpretable proxy, revealing structural changes in the model's decision boundary and its underlying rationale. We operationalize this analysis within a three-layer framework that synergistically combines insights from the data layer (distributional shifts), the model layer (prediction disagreement), and our proposed explanation layer. We show that such holistic view allows for a more comprehensive diagnosis of drift, making it possible to distinguish between different root causes, such as a spatial data shift versus a re-labeling of concepts.