Efficient and High-quality Recommendations via Momentum-incorporated Parallel Stochastic Gradient Descent-Based Learning

doi:10.1109/JAS.2020.1003396

	Efficient and High-quality Recommendations via Momentum-incorporated Parallel Stochastic Gradient Descent-Based Learning
	Luo, Xin 4,5; Qin, Wen 6,7; Dong, Ani 8; Sedraoui, Khaled 1,3; Zhou, MengChu 2,3
刊名	IEEE-CAA JOURNAL OF AUTOMATICA SINICA
	2021-02-01
卷号	8 期号:2 页码:402-411
关键词	Big data industrial application industrial data latent factor analysis machine learning parallel algorithm recommender system (RS) stochastic gradient descent (SGD)
ISSN号	2329-9266
DOI	10.1109/JAS.2020.1003396
通讯作者	Zhou, MengChu(zhou@njit.edu)
英文摘要	A recommender system (RS) relying on latent factor analysis usually adopts stochastic gradient descent (SGD) as its learning algorithm. However, owing to its serial mechanism, an SGD algorithm suffers from low efficiency and scalability when handling large-scale industrial problems. Aiming at addressing this issue, this study proposes a momentum-incorporated parallel stochastic gradient descent (MPSGD) algorithm, whose main idea is two-fold: a) implementing parallelization via a novel datasplitting strategy, and b) accelerating convergence rate by integrating momentum effects into its training process. With it, an MPSGD-based latent factor (MLF) model is achieved, which is capable of performing efficient and high-quality recommendations. Experimental results on four high-dimensional and sparse matrices generated by industrial RS indicate that owing to an MPSGD algorithm, an MLF model outperforms the existing state-of-the-art ones in both computational efficiency and scalability.
资助项目	National Natural Science Foundation of China[61772493] ; King Abdulaziz University[RG-48-135-40] ; Guangdong Province Universities and College Pearl River Scholar Funded Scheme (2019) ; Natural Science Foundation of Chongqing[cstc2019jcyjjqX0013]
WOS研究方向	Automation & Control Systems
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000607401900008
内容类型	期刊论文
源URL	[http://119.78.100.138/handle/2HOD01W0/12834]
专题	中国科学院重庆绿色智能技术研究院
通讯作者	Zhou, MengChu
作者单位	1.King Abdulaziz Univ, Dept Elect & Comp Engn, Fac Engn, Jeddah 21481, Saudi Arabia 2.New Jersey Inst Technol, Dept Elect & Comp Engn, Newark, NJ 07102 USA 3.King Abdulaziz Univ, Ctr Res Excellence Renewable Energy & Power Syst, Jeddah 21481, Saudi Arabia 4.Dongguan Univ Technol, Sch Comp Sci & Technol, Dongguan 523808, Peoples R China 5.Cloudwalk, Dept Big Data Anal Tech, Hengrui Chongqing Artificial Intelligence Res Ctr, Chongqing 401331, Peoples R China 6.Chongqing Univ Posts & Telecommun, Sch Comp Sci & Technol, Chongqing 400065, Peoples R China 7.Chinese Acad Sci, Chongqing Inst Green & Intelligent Technol, Chongqing Engn Res Ctr Big Data Applicat Smart Ci, Chongqing 400714, Peoples R China 8.Dongguan Univ Technol, City Coll, Dept Comp & Informat Sci, Dongguan 523419, Peoples R China
推荐引用方式 GB/T 7714	Luo, Xin,Qin, Wen,Dong, Ani,et al. Efficient and High-quality Recommendations via Momentum-incorporated Parallel Stochastic Gradient Descent-Based Learning[J]. IEEE-CAA JOURNAL OF AUTOMATICA SINICA,2021,8(2):402-411.
APA	Luo, Xin,Qin, Wen,Dong, Ani,Sedraoui, Khaled,&Zhou, MengChu.(2021).Efficient and High-quality Recommendations via Momentum-incorporated Parallel Stochastic Gradient Descent-Based Learning.IEEE-CAA JOURNAL OF AUTOMATICA SINICA,8(2),402-411.
MLA	Luo, Xin,et al."Efficient and High-quality Recommendations via Momentum-incorporated Parallel Stochastic Gradient Descent-Based Learning".IEEE-CAA JOURNAL OF AUTOMATICA SINICA 8.2(2021):402-411.