Fast and Slow Streams for Online Time Series Forecasting Without Information Leakage
Keywords: online time series forecasting, concept drift, online learning
TL;DR: Redefined the setting of online time series forecasting to prevent information leakage and proposed a model-agnostic framework for this setting.
Abstract: Current research in online time series forecasting suffers from information leakage: models predict and then evaluate on historical time steps that have been backpropagated for parameter updates. This setting also misaligns with the real-world conception of forecasting, which typically emphasizes looking ahead and anticipating future uncertainties. This paper redefines online time series forecasting to focus on predicting unknown future steps and evaluates performance solely based on these predictions. Following this new setting, challenges arise in leveraging incomplete pairs of ground truth and prediction for backpropagation, as well as generalizing accurate information without overfitting to noises from recent data streams. To address these challenges, we propose a novel dual-stream framework for online forecasting (DSOF): a slow stream that updates with complete data using experience replay, and a fast stream that adapts to recent data through temporal difference learning. This dual-stream approach updates a teacher-student model learned through a residual learning strategy, generating predictions in a coarse-to-fine manner. Extensive experiments demonstrate its improvement in forecasting performance in changing environments.
Primary Area: learning on time series and dynamical systems
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 9019
Loading