2019

Block based Singular Value Decomposition approach to matrix factorization for recommender systems

Bhavana, Prasad, Kumar, Vikas, Padmanabhan, Vineet

Understand

With the abundance of data in recent years, interesting challenges are posed in the area of recommender systems.

Producing high quality recommendations with scalability and performance is the need of the hour.
Singular Value Decomposition(SVD) based recommendation algorithms have been leveraged to produce better results.
In this paper, we extend the SVD technique further for scalability and performance in the context of 1) multi-threading 2) multiple computational units (with the use of Graphical Processing Units) and 3) distributed computation.

Built on

Indexing by latent semantic analysis
Scott Deerwester, Susan T. Dumais, George W. Furnas, Thomas K. Landauer, and Richard Harshman · 1990
Earlier work this paper cites.
Using linear algebra for intelligent information retrieval
M. W. Berry, S.T. Dumais, and G.W. O’Brien · 1995
Earlier work this paper cites.
Incremental singular value decomposition algorithms for highly scalable recommender systems
Badrul Sarwar, George Karypis, Joseph Konstan, and John Riedl · 2002
Earlier work this paper cites.
Dimensionality reduction in higher-order signal processing and rank-(r1,r2,…,rn) reduction in multilinear algebra
Lieven De Lathauwer and Joos Vandewalle · 2004
Earlier work this paper cites.
Incremental kernel svd for face recognition with image sets
Tat-Jun Chin, K. Schindler, and D. Suter · 2006
Earlier work this paper cites.
Probabilistic matrix factorization
Ruslan Salakhutdinov and Andriy Mnih · 2007
Earlier work this paper cites.
Incremental learning for robust visual tracking
David A. Ross, Jongwoo Lim, Ruei-Sung Lin, and Ming-Hsuan Yang · 2008
Earlier work this paper cites.
1 the bellkor solution to the netflix grand prize, 2009
Yehuda Koren · 2009
Earlier work this paper cites.
A survey of collaborative filtering techniques
Xiaoyuan Su and Taghi M Khoshgoftaar · 2009
Earlier work this paper cites.
Large-scale matrix factorization with distributed stochastic gradient descent
Rainer Gemulla, Erik Nijkamp, Peter J. Haas, and Yannis Sismanis · 2011
Earlier work this paper cites.

Similar

Divide-and-conquer matrix factorization
Lester W. Mackey, Michael I. Jordan, and Ameet Talwalkar · 2011
Cited alongside, same era.
A very fast algorithm for matrix factorization
Vladimir Nikulin, Tian-Hsiang Huang, Shu-Kay Ng, Suren I. Rathnayake, and Geoffrey J. McLachlan · 2011
Cited alongside, same era.
Hogwild: A lock-free approach to parallelizing stochastic gradient descent
Benjamin Recht, Christopher Ré, Stephen J. Wright, and Feng Niu · 2011
Cited alongside, same era.
Gpu-accelerated non-negative matrix factorization for text mining
Volodymyr Kysenko, Karl Rupp, Oleksandr Marchenko, Siegfried Selberherr, and Anatoly Anisimov · 2012
Cited alongside, same era.
Scalable coordinate descent approaches to parallel matrix factorization for recommender systems
Hsiang-Fu Yu, Cho-Jui Hsieh, Si Si, and Inderjit S. Dhillon · 2012
Cited alongside, same era.
Sparkler: supporting large-scale matrix factorization
Boduo Li, Sandeep Tata, and Yannis Sismanis · 2013
Cited alongside, same era.
Split-and-combine singular value decomposition for large-scale matrix
Jengnan Tzeng · 2013
Cited alongside, same era.
Localized matrix factorization for recommendation based on matrix block diagonal forms
Yongfeng Zhang, Min Zhang, Yiqun Liu, Shaoping Ma, and Shi Feng · 2013
Cited alongside, same era.
A fast parallel stochastic gradient descent for matrix factorization in shared memory systems
Yong Zhuang, Wei-Sheng Chin, Yu-Chin Juan, and Chih-Jen Lin · 2013
Cited alongside, same era.
Nomad: Non-locking, stochastic multi-machine algorithm for asynchronous and decentralized matrix completion
Hyokun Yun, Hsiang-Fu Yu, Cho-Jui Hsieh, S. V. N. Vishwanathan, and Inderjit Dhillon
Cited in the paper.

Then

Factorbird - a parameter server approach to distributed matrix factorization
Original
Sebastian Schelter, Venu Satuluri, and Reza Zadeh · 2014
Later among the works it cites.
A learning-rate schedule for stochastic gradient methods to matrix factorization
Wei-Sheng Chin, Yong Zhuang, Yu-Chin Juan, and Chih-Jen Lin · 2015
Later among the works it cites.
Fast and robust parallel SGD matrix factorization
Jinoh Oh, Wook-Shin Han, Hwanjo Yu, and Xiaoqian Jiang · 2015
Later among the works it cites.
Block kernel non-negative matrix factorization and its application to face recognition
W. Chen, Y. Li, B. Pan, and C. Xu · 2016
Later among the works it cites.
nmfgpu4r: Gpu-accelerated computation of the non-negative matrix factorization using cuda capable hardware
Sven Koitka and Christoph M Friedrich · 2016
Later among the works it cites.
Dc-nmf: Non-negative matrix factorization based on divide-and-conquer for fast clustering and topic modeling
Rundong Du, Da Kuang, Barry Drake, and Haesun Park · 2017
Later among the works it cites.
Collaborative filtering using multiple binary maximum margin matrix factorizations
Vikas Kumar, Arun K. Pujari, Sandeep Kumar Sahu, Venkateswara Rao Kagita, and Vineet Padmanabhan · 2017
Later among the works it cites.
Proximal maximum margin matrix factorization for collaborative filtering
Vikas Kumar, Arun K. Pujari, Sandeep Kumar Sahu, Venkateswara Rao Kagita, and Vineet Padmanabhan · 2017
Later among the works it cites.
Matrix factorization on gpus with memory optimization and approximate computing
Wei Tan, Shiyu Chang, Liana L. Fong, Cheng Li, Zijun Wang, and Liang Cao · 2018
Later among the works it cites.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…

Reading the bibliography…