REFOL: Resource-Efficient Federated Online Learning for Traffic Flow Forecasting

doi:10.1109/TITS.2024.3510913

	REFOL: Resource-Efficient Federated Online Learning for Traffic Flow Forecasting
	Liu, Qingxiang 1,2; Sun, Sheng 1; Liang, Yuxuan 3,4; Xu, Xiaolong 5,6; Liu, Min 1,7; Bilal, Muhammad 8; Wang, Yuwei 1; Li, Xujing 1,2; Zheng, Yu 9,10
	2024-12-17
发表期刊	IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS
ISSN	1524-9050
页码	16
摘要	Multiple federated learning (FL) methods are proposed for traffic flow forecasting (TFF) to avoid heavy-transmission and privacy-leaking concerns resulting from the disclosure of raw data in centralized methods. However, these FL methods adopt offline learning which may yield subpar performance, when concept drift occurs, i.e., distributions of historical and future data vary. Online learning can detect concept drift during model training, thus more applicable to TFF. Nevertheless, the existing federated online learning method for TFF fails to efficiently solve the concept drift problem and causes tremendous computing and communication overhead. Therefore, we propose a novel method named Resource-Efficient Federated Online Learning (REFOL) for TFF, which guarantees prediction performance in a communication-lightweight and computation-efficient way. Specifically, we design a data-driven client participation mechanism to detect the occurrence of concept drift and determine clients' participation necessity. Subsequently, we propose an adaptive online optimization strategy, which guarantees prediction performance and meanwhile avoids meaningless model updates. Then, a graph convolution-based model aggregation mechanism is designed, aiming to assess participants' contribution based on spatial correlation without importing extra communication and computing consumption on clients. Finally, we conduct extensive experiments on real-world datasets to demonstrate the superiority of REFOL in terms of prediction improvement and resource economization.
关键词	Predictive models Concept drift Correlation Forecasting Computational modeling Data models Optimization Adaptation models Urban areas Training Traffic flow forecasting federated learning concept drift online learning graph convolution
DOI	10.1109/TITS.2024.3510913
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program of China[2021YFB2900102] ; National Natural Science Foundation of China[62472410] ; National Natural Science Foundation of China[62072436]
WOS研究方向	Engineering ; Transportation
WOS类目	Engineering, Civil ; Engineering, Electrical & Electronic ; Transportation Science & Technology
WOS记录号	WOS:001381467400001
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/41065
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Liu, Min
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100190, Peoples R China 3.Hong Kong Univ Sci & Technol Guangzhou, INTR Thrust, Guangzhou 510000, Peoples R China 4.Hong Kong Univ Sci & Technol Guangzhou, DSA Thrust, Guangzhou 510000, Peoples R China 5.Nanjing Univ Informat Sci & Technol, Sch Software, Nanjing 210044, Peoples R China 6.Nanjing Univ Informat Sci & Technol, Jiangsu Collaborat Innovat Ctr Atmospher Environm, Nanjing 210044, Peoples R China 7.Zhongguancun Lab, Beijing 100094, Peoples R China 8.Univ Lancaster, Sch Comp & Commun, Lancaster LA1 4WA, England 9.JD Intelligent Cities Res, Beijing 102300, Peoples R China 10.JD Technol, JD iCity, Beijing 102300, Peoples R China
推荐引用方式 GB/T 7714	Liu, Qingxiang,Sun, Sheng,Liang, Yuxuan,et al. REFOL: Resource-Efficient Federated Online Learning for Traffic Flow Forecasting[J]. IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS,2024:16.
APA	Liu, Qingxiang.,Sun, Sheng.,Liang, Yuxuan.,Xu, Xiaolong.,Liu, Min.,...&Zheng, Yu.(2024).REFOL: Resource-Efficient Federated Online Learning for Traffic Flow Forecasting.IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS,16.
MLA	Liu, Qingxiang,et al."REFOL: Resource-Efficient Federated Online Learning for Traffic Flow Forecasting".IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS (2024):16.