Enregistré dans:
Détails bibliographiques
Auteurs principaux: Lin, Shengsheng, Lin, Weiwei, Hu, Xinyi, Wu, Wentai, Mo, Ruichao, Zhong, Haocheng
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2409.18479
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917803479531520
author Lin, Shengsheng
Lin, Weiwei
Hu, Xinyi
Wu, Wentai
Mo, Ruichao
Zhong, Haocheng
author_facet Lin, Shengsheng
Lin, Weiwei
Hu, Xinyi
Wu, Wentai
Mo, Ruichao
Zhong, Haocheng
contents The stable periodic patterns present in time series data serve as the foundation for conducting long-horizon forecasts. In this paper, we pioneer the exploration of explicitly modeling this periodicity to enhance the performance of models in long-term time series forecasting (LTSF) tasks. Specifically, we introduce the Residual Cycle Forecasting (RCF) technique, which utilizes learnable recurrent cycles to model the inherent periodic patterns within sequences, and then performs predictions on the residual components of the modeled cycles. Combining RCF with a Linear layer or a shallow MLP forms the simple yet powerful method proposed in this paper, called CycleNet. CycleNet achieves state-of-the-art prediction accuracy in multiple domains including electricity, weather, and energy, while offering significant efficiency advantages by reducing over 90% of the required parameter quantity. Furthermore, as a novel plug-and-play technique, the RCF can also significantly improve the prediction accuracy of existing models, including PatchTST and iTransformer. The source code is available at: https://github.com/ACAT-SCUT/CycleNet.
format Preprint
id arxiv_https___arxiv_org_abs_2409_18479
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle CycleNet: Enhancing Time Series Forecasting through Modeling Periodic Patterns
Lin, Shengsheng
Lin, Weiwei
Hu, Xinyi
Wu, Wentai
Mo, Ruichao
Zhong, Haocheng
Machine Learning
The stable periodic patterns present in time series data serve as the foundation for conducting long-horizon forecasts. In this paper, we pioneer the exploration of explicitly modeling this periodicity to enhance the performance of models in long-term time series forecasting (LTSF) tasks. Specifically, we introduce the Residual Cycle Forecasting (RCF) technique, which utilizes learnable recurrent cycles to model the inherent periodic patterns within sequences, and then performs predictions on the residual components of the modeled cycles. Combining RCF with a Linear layer or a shallow MLP forms the simple yet powerful method proposed in this paper, called CycleNet. CycleNet achieves state-of-the-art prediction accuracy in multiple domains including electricity, weather, and energy, while offering significant efficiency advantages by reducing over 90% of the required parameter quantity. Furthermore, as a novel plug-and-play technique, the RCF can also significantly improve the prediction accuracy of existing models, including PatchTST and iTransformer. The source code is available at: https://github.com/ACAT-SCUT/CycleNet.
title CycleNet: Enhancing Time Series Forecasting through Modeling Periodic Patterns
topic Machine Learning
url https://arxiv.org/abs/2409.18479