Phase-Driven Domain Generalizable Learning For Nonstationary Time Series Classification

Payal Mohapatra, Lixu Wang, Qi Zhu

Keywords: Time Series; Domain Generalization; Machine Learning for Time Series data;

TL;DR: Incorporating phase information across augmentation, feature extraction, and residual connections stages enhances generalization performance in non-stationary time-series applications.

Abstract: Monitoring and recognizing patterns in continuous sensing data is crucial for many practical applications. These real-world time-series data are often nonstationary, characterized by varying statistical and spectral properties over time. This poses a significant challenge in developing learning models that can effectively generalize across different distributions. In this work, based on our observation that nonstationary statistics for time-series classification tasks are intrinsically linked to the phase information, we propose a time-series domain generalization framework, PhASER. It consists of three novel elements: 1) Hilbert transform-based phase augmentation that diversifies non-stationarity while preserving discriminatory semantics, 2) separate magnitude-phase encoding by viewing time-varying magnitude and phase as independent modalities, and 3) phase-residual feature broadcasting by incorporating phase with a novel residual connection for inherent regularization to enhance distribution invariant learning. Extensive evaluation on 5 datasets from sleep-stage classification, human activity recognition, and gesture recognition against 12 state-of-the-art baseline methods demonstrate that PhASER consistently outperforms the best baselines by an average of 5% and up to 13% in some cases. Moreover, PhASER’s principles can also be applied broadly to boost the generalizability of existing time-series classification models.

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: 5718