A Fast Parallel Stochastic Gradient Method for Matrix Factorization in Shared Memory Systems

被引：73

作者：

Chin, Wei-Sheng ^{[1
]}

Zhuang, Yong ^{[1
]}

Juan, Yu-Chin ^{[1
]}

Lin, Chih-Jen ^{[1
]}

机构：

[1] Natl Taiwan Univ, Dept Comp Sci, Taipei, Taiwan

来源：

ACM TRANSACTIONS ON INTELLIGENT SYSTEMS AND TECHNOLOGY | 2015年 / 6卷 / 01期

关键词：

Parallel and Vector Implementations; Recommender system; matrix factorization; stochastic gradient descent; parallel computing; shared memory algorithm;

D O I：

10.1145/2668133

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Matrix factorization is known to be an effective method for recommender systems that are given only the ratings from users to items. Currently, stochastic gradient (SG) method is one of the most popular algorithms for matrix factorization. However, as a sequential approach, SG is difficult to be parallelized for handling web-scale problems. In this article, we develop a fast parallel SG method, FPSG, for shared memory systems. By dramatically reducing the cache-miss rate and carefully addressing the load balance of threads, FPSG is more efficient than state-of-the-art parallel algorithms for matrix factorization.

引用

页数：24

共 50 条

[21] Parallel matrix factorization for recommender systems
Yu, Hsiang-Fu
Hsieh, Cho-Jui
Si, Si
Dhillon, Inderjit S.
[J]. KNOWLEDGE AND INFORMATION SYSTEMS, 2014, 41 (03) : 793 - 819
[22] An Efficient Approach of GPU-accelerated Stochastic Gradient Descent Method for Matrix Factorization
Li, Feng
Ye, Yunming
Li, Xutao
[J]. JOURNAL OF INTERNET TECHNOLOGY, 2019, 20 (04): : 1087 - 1097
[23] A NEW APPROACH OF GPU-ACCELERATED STOCHASTIC GRADIENT DESCENT METHOD FOR MATRIX FACTORIZATION
Li, Feng
Ye, Yunming
Li, Xutao
Lu, Jiajie
[J]. INTERNATIONAL JOURNAL OF INNOVATIVE COMPUTING INFORMATION AND CONTROL, 2019, 15 (02): : 697 - 711
[24] Parallel sparse LU factorization in a shared memory computing environment
Schenk, O
Gärtner, K
Fichtner, W
[J]. INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED PROCESSING TECHNIQUES AND APPLICATIONS, VOLS I-IV, PROCEEDINGS, 1998, : 907 - 914
[25] PARALLEL SPARSE QR FACTORIZATION ON SHARED-MEMORY ARCHITECTURES
MATSTOMS, P
[J]. PARALLEL COMPUTING, 1995, 21 (03) : 473 - 486
[26] SPLIT GRADIENT METHOD FOR NONNEGATIVE MATRIX FACTORIZATION
Lanteri, Henri
Theys, Celine
Richard, Cedric
Fevotte, Cedric
[J]. 18TH EUROPEAN SIGNAL PROCESSING CONFERENCE (EUSIPCO-2010), 2010, : 1199 - 1203
[27] The Convergence of Stochastic Gradient Descent in Asynchronous Shared Memory
Alistarh, Dan
De Sa, Christopher
Konstantinov, Nikola
[J]. PODC'18: PROCEEDINGS OF THE 2018 ACM SYMPOSIUM ON PRINCIPLES OF DISTRIBUTED COMPUTING, 2018, : 169 - 177
[28] QR FACTORIZATION OF A DENSE MATRIX ON A SHARED-MEMORY MULTIPROCESSOR
CHU, E
GEORGE, A
[J]. PARALLEL COMPUTING, 1989, 11 (01) : 55 - 71
[29] Accelerating Stochastic Gradient Descent Based Matrix Factorization on FPGA
Zhou, Shijie
Kannan, Rajgopal
Prasanna, Viktor K.
[J]. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 2020, 31 (08) : 1897 - 1911
[30] Parallel QR factorization for hybrid message passing/shared memory operation
Dunn, IN
Meyer, GGL
[J]. JOURNAL OF THE FRANKLIN INSTITUTE-ENGINEERING AND APPLIED MATHEMATICS, 2001, 338 (05): : 601 - 613

← 1 2 3 4 5 →