A parallelized, momentum-incorporated stochastic gradient descent scheme for latent factor analysis on high-dimensional and sparse matrices from recommender systems

doi:10.1109/SMC.2019.8914671

	A parallelized, momentum-incorporated stochastic gradient descent scheme for latent factor analysis on high-dimensional and sparse matrices from recommender systems
	Qin, Wen 1; Wu, Hao 1; Lai, Qingkuan 1; Wang, Chaobin 2
	2019
会议日期	October 6, 2019 - October 9, 2019
会议地点	Bari, Italy
DOI	10.1109/SMC.2019.8914671
页码	1744-1749
英文摘要	High-dimensional and sparse (HiDS) matrices are commonly encountered in many big-data-related industrial applications like recommender systems. Latent factor (LF) analysis via stochastic gradient descent (SGD) is greatly efficient in discovering latent patterns from them. However, as a sequential algorithm, SGD suffers considerable time cost and low scalability when handling large-scale problems. To address these issues, this study proposes parallelized, momentum-incorporated stochastic gradient descent (PMSGD) scheme, which incorporates momentum effects into an SGD scheme as well as implementing its parallelization via careful data splitting. Based on a PMSGD method, we achieve a PMSGD-based LF (PLF) model to execute fast LF analysis on HiDS matrices from a recommender system. Experimental results on two HiDS matrices arising from industrial applications indicate that owing to the careful design of PMSGD, a PLF model outperforms state-of-the-art parallel LF models significantly in terms of computational efficiency. © 2019 IEEE.
会议录	2019 IEEE International Conference on Systems, Man and Cybernetics, SMC 2019
语种	英语
ISSN号	1062922X
内容类型	会议论文
源URL	[http://119.78.100.138/handle/2HOD01W0/9796]
专题	中国科学院重庆绿色智能技术研究院
作者单位	1.Computer School of China West Normal University, Nanchong, Sichuan; 637002, China; 2.Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences, Chongqing Key Laboratory of Big Data and Intelligent Computing, Chongqing; 400714, China
推荐引用方式 GB/T 7714	Qin, Wen,Wu, Hao,Lai, Qingkuan,et al. A parallelized, momentum-incorporated stochastic gradient descent scheme for latent factor analysis on high-dimensional and sparse matrices from recommender systems[C]. 见:. Bari, Italy. October 6, 2019 - October 9, 2019.