Explaining Time Series Classification Predictions via Causal Attributions

TL;DR

This paper introduces a causal attribution method for time series classification that compares causal and associational explanations, revealing important differences and emphasizing the importance of causal analysis over purely associational methods.

Contribution

The paper presents a novel, model-agnostic causal attribution approach for time series classification, utilizing diffusion models for counterfactual estimation and comparing it with traditional associational methods.

Findings

Causal and associational attributions often differ significantly.

The proposed method provides deeper insights into model decision-making.

Causal attributions reveal limitations of associational explanations.

Abstract

Despite the excelling performance of machine learning models, understanding their decisions remains a long-standing goal. Although commonly used attribution methods from explainable AI attempt to address this issue, they typically rely on associational rather than causal relationships. In this study, within the context of time series classification, we introduce a novel model-agnostic attribution method to assess the causal effect of concepts i.e., predefined segments within a time series, on classification outcomes. Our approach compares these causal attributions with closely related associational attributions, both theoretically and empirically. To estimate counterfactual outcomes, we use state-of-the-art diffusion models backed by state space models. We demonstrate the insights gained by our approach for a diverse set of qualitatively different time series classification tasks. Although causal and associational attributions might often share some similarities, in all cases they differ in important details, underscoring the risks associated with drawing causal conclusions from associational data alone. We believe that the proposed approach is also widely applicable in other domains to shed some light on the limits of associational attributions.