EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

TimeHome: Heterogeneous Mixture-of-Experts for Time-Series Foundation Model

Tao Zhang, Xiaobo Wu, Xingguo Li, Donghua Wu

Remote Sensing · 03.08.2026 · 18 (15) · 2533

DOI: 10.3390/rs18152533 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 03.08.2026. Verfasst von Tao Zhang und 3 weiteren Autor:innen. Die Arbeit nennt 79 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Zhejiang University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Time-series analysis is important for various scientific and industrial fields, such as remote sensing where observations may be disturbed by clouds, have irregular revisits and different sensors. Current transformer-based approaches have enhanced the modeling of distant time relationships, but they are still constrained by specific task designs and poor adaptability to various time-series patterns. To solve these problems, we present TimeHome, a universal sparse transformer basic model for handling heterogeneous time series. TimeHome incorporates a Heterogeneous Mixture-of-Experts (H-MoE) component, where different expert types are chosen dynamically based on a low-rank temperature-controlled gating mechanism to fit various sequence features. Moreover, a hybrid local–global attention mechanism is designed to consider both short-term variations and long-distance correlations, while specific heads are used for unified prediction, missing value estimation and abnormal event detection. TimeHome is pretrained on TS-200B, a huge database of time series including various temporal patterns from different domains. Comprehensive tests on several benchmark datasets and remote sensing extended evaluations show that TimeHome performs well in long-term prediction, missing value replacement and abnormal event detection. The model also exhibits good zero-shot adaptation ability and fast inference speed by adjusting experts dynamically. The source code and pre-training data will be released publicly.

Förderung

  • Zhejiang University

Zitieren

APA

Zhang, T., Wu, X., Li, X., & Wu, D. (2026). TimeHome: Heterogeneous Mixture-of-Experts for Time-Series Foundation Model. Remote Sensing, 18(15), 2533. https://doi.org/10.3390/rs18152533

Vancouver

Zhang T, Wu X, Li X, Wu D. TimeHome: Heterogeneous Mixture-of-Experts for Time-Series Foundation Model. Remote Sensing. 2026;18(15):2533. doi:10.3390/rs18152533

BibTeX
@article{Zhang2026,
  author = {Zhang, Tao and Wu, Xiaobo and Li, Xingguo and Wu, Donghua},
  title = {TimeHome: Heterogeneous Mixture-of-Experts for Time-Series Foundation Model},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {15},
  pages = {2533},
  doi = {10.3390/rs18152533}
}
Kostenlos starten →