Publikation
Right on Time: Revising Time Series Models by Constraining their Explanations
Maurice Kraus; David Steinmann; Antonia Wüst; Andre Kokozinski; Kristian Kersting
In: Computing Research Repository eprint Journal (CoRR), Vol. abs/2402.12921, Pages 1-31, arXiv, 2024.
Zusammenfassung
Deep time series models often suffer from reliability issues
due to their tendency to rely on spurious correlations, leading to incor-
rect 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 interac-
tion strategy is crucial for effectively addressing shortcuts in time series
datasets. We empirically demonstrate the effectiveness of RioT in guid-
ing models toward more reliable decision-making across popular time se-
ries classification and forecasting datasets, as well as our newly recorded
dataset with naturally occuring shortcuts, P2S, collected from a real
mechanical production line.