Right on Time: Revising Time Series Models by Constraining their Explanations

TL;DR

This paper introduces RioT, a method that improves the reliability of deep time series models by constraining their explanations in both time and frequency domains to avoid reliance on spurious correlations.

Contribution

RioT is a novel approach that interacts with model explanations across time and frequency domains to reduce shortcuts in time series models.

Findings

RioT effectively guides models toward more reliable decisions.

Demonstrated on multiple datasets including real-world production line data.

Improves model robustness against spurious correlations.

Abstract

Deep time series models often suffer from reliability issues due to their tendency to rely on spurious correlations, leading to incorrect predictions. To mitigate such shortcuts and prevent "Clever-Hans" moments in time series models, we introduce Right on Time (RioT), a novel method that enables interacting with model explanations across both the time and frequency domains. By incorporating feedback on explanations in both domains, RioT constrains the model, steering it away from annotated spurious correlations. This dual-domain interaction strategy is crucial for effectively addressing shortcuts in time series datasets. We empirically demonstrate the effectiveness of RioT in guiding models toward more reliable decision-making across popular time series classification and forecasting datasets, as well as our newly recorded dataset with naturally occuring shortcuts, P2S, collected from a real mechanical production line.