EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Transfer Learning Based on N-BEATS in Forecasting Univariate Time Series

Minwoo Lee, Youngmi Lee, Gwangsu Kim

IEEE Access · 2026 · 14 · 45191-45212

DOI: 10.1109/access.2026.3675559 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Access, 2026. Verfasst von Minwoo Lee und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 62 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Global—Learning and Academic Research Institution for Master’s and Ph.D. students, and Postdocs (LAMP) Program of the National Research Foundation of Korea, Ministry of Education, NRF.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Transfer learning has emerged as a powerful technique in machine learning, wherein knowledge acquired from one task or dataset is leveraged to improve performance on related tasks and datasets. Although univariate time series forecasting plays a pivotal role in numerous real-world applications, the adoption of transfer learning in this domain has been relatively underexplored. This paper proposes a new transfer learning approach for univariate time series forecasting, combining the Neural Basis Expansion Analysis for interpretable Time Series forecasting (N-BEATS) and Multi-Layer Perceptron (MLP). Furthermore, the Boltzmann-weighted ensemble is proposed for prediction, serving as an advanced inference method that improves forecasting accuracy without incurring significant computational overhead. We constructed a pre-trained model for knowledge transfer. Then, we fine-tuned it to predict time series data from seven domains (temperature, solar radiation, electricity demand, air pollution, wind speed, traffic volume, and coin price). Our experimental results show that transfer learning enhances predictive performance, and investigations into additional layers reveal that MLP architectures are superior to long short-term memory and original transformer architectures in terms of Root Mean Square Error (RMSE), whereas the advanced time-series transformer PatchTST shows potential in transfer learning. Additionally, we conducted various experiments for qualitative analyses. The results of experiments consistently demonstrate the superior performance of transfer learning for univariate time series forecasting, offering valuable insights into designing more robust deep learning solutions in data-driven environments.

Förderung

  • Global—Learning and Academic Research Institution for Master’s and Ph.D. students, and Postdocs (LAMP) Program of the National Research Foundation of Korea
  • Ministry of Education
  • NRF
  • Korean Government

Zitieren

APA

Lee, M., Lee, Y., & Kim, G. (2026). Transfer Learning Based on N-BEATS in Forecasting Univariate Time Series. IEEE Access, 14, 45191-45212. https://doi.org/10.1109/access.2026.3675559

Vancouver

Lee M, Lee Y, Kim G. Transfer Learning Based on N-BEATS in Forecasting Univariate Time Series. IEEE Access. 2026;14:45191-45212. doi:10.1109/access.2026.3675559

BibTeX
@article{Lee2026,
  author = {Lee, Minwoo and Lee, Youngmi and Kim, Gwangsu},
  title = {Transfer Learning Based on N-BEATS in Forecasting Univariate Time Series},
  journal = {IEEE Access},
  year = {2026},
  volume = {14},
  pages = {45191--45212},
  doi = {10.1109/access.2026.3675559}
}
Kostenlos starten →