Guardado en:
Detalles Bibliográficos
Autores principales: Liu, Yangyou, Shao, Zezhi, Chen, Xinyu, Chen, Hu, Wang, Fei, Wu, Yuankai
Formato: Preprint
Publicado: 2026
Materias:
Acceso en línea:https://arxiv.org/abs/2605.16793
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866913149309943808
author Liu, Yangyou
Shao, Zezhi
Chen, Xinyu
Chen, Hu
Wang, Fei
Wu, Yuankai
author_facet Liu, Yangyou
Shao, Zezhi
Chen, Xinyu
Chen, Hu
Wang, Fei
Wu, Yuankai
contents Time series forecasting under non-stationarity faces a fundamental tension between capturing stable representations and adapting to distribution shifts. Existing methods implicitly rely on static historical assumptions, leading to a critical failure mode we term Phase Amnesia, where models become blind to the evolving global context. To resolve this, we formalize non-stationary dynamics through three physical hypotheses: wold decomposition, dynamical phase evolution, and heteroscedastic manifold generation. These principles inspire PULSE, a physics-informed, plug-and-play framework adopting a Disentangle--Evolve--Simulate design philosophy. Specifically, PULSE utilizes phase-anchored disentanglement to resolve optimization interference caused by dominant trends, employs a Phase Router to actively generate future trajectories, and introduces Statistic-Aware Mixup (SAM) to ensure robustness against out-of-distribution volatility. Empirically, PULSE enables a simple MLP backbone to achieve state-of-the-art or highly competitive performance across 12 real-world benchmarks. This validates that a correct physics-informed inductive bias is far more critical than raw architectural complexity for non-stationary forecasting. The code is available at: https://github.com/Gemost/PULSE.
format Preprint
id arxiv_https___arxiv_org_abs_2605_16793
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle PULSE: Generative Phase Evolution for Non-Stationary Time Series Forecasting
Liu, Yangyou
Shao, Zezhi
Chen, Xinyu
Chen, Hu
Wang, Fei
Wu, Yuankai
Machine Learning
Time series forecasting under non-stationarity faces a fundamental tension between capturing stable representations and adapting to distribution shifts. Existing methods implicitly rely on static historical assumptions, leading to a critical failure mode we term Phase Amnesia, where models become blind to the evolving global context. To resolve this, we formalize non-stationary dynamics through three physical hypotheses: wold decomposition, dynamical phase evolution, and heteroscedastic manifold generation. These principles inspire PULSE, a physics-informed, plug-and-play framework adopting a Disentangle--Evolve--Simulate design philosophy. Specifically, PULSE utilizes phase-anchored disentanglement to resolve optimization interference caused by dominant trends, employs a Phase Router to actively generate future trajectories, and introduces Statistic-Aware Mixup (SAM) to ensure robustness against out-of-distribution volatility. Empirically, PULSE enables a simple MLP backbone to achieve state-of-the-art or highly competitive performance across 12 real-world benchmarks. This validates that a correct physics-informed inductive bias is far more critical than raw architectural complexity for non-stationary forecasting. The code is available at: https://github.com/Gemost/PULSE.
title PULSE: Generative Phase Evolution for Non-Stationary Time Series Forecasting
topic Machine Learning
url https://arxiv.org/abs/2605.16793