Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/26395
Title: | Distributed Bayesian Probabilistic Matrix Factorization | Authors: | Aa, Tom Vander Chakroun, Imen HABER, Tom |
Issue Date: | 2017 | Publisher: | Elsevier Science BV | Source: | Koumoutsakos, Petros; Lees, Michael; Krzhizhanovskaya, Valeria; Dongarra, Jack; Sloot, Peter M. A. (Ed.). International Conference on Computational Science, ICCS 2017, 12-14 June 2017, Zurich, Switzerland, Elsevier Science BV,p. 1030-1039 | Series/Report: | Procedia Computer Science | Series/Report no.: | 108 | Abstract: | Using the matrix factorization technique in machine learning is very common mainly in areas like recommender systems. Despite its high prediction accuracy and its ability to avoid over-fitting of the data, the Bayesian Probabilistic Matrix Factorization algorithm (BPMF) has not been widely used on large scale data because of the prohibitive cost. In this paper, we propose a distributed high-performance parallel implementation of the BPMF using Gibbs sampling on shared and distributed architectures. We show by using efficient load balancing using work stealing on a single node, and by using asynchronous communication in the distributed version we beat state of the art implementations. (C) 2017 The Authors. Published by Elsevier B.V. | Notes: | [Aa, Tom Vander; Chakroun, Imen] IMEC, Exascience Lab, Kapeldreef 75, B-3001 Leuven, Belgium. [Haber, Tom] Expertise Ctr Digital Media, Wetenschapspk 2, B-3590 Diepenbeek, Belgium. | Keywords: | Probabilistic matrix factorization algorithm; Collaborative filtering; Machine learning; PGAS; multi-core;probabilistic matrix factorization algorithm; Collaborative filtering; machine learning; PGAS; multi-core | Document URI: | http://hdl.handle.net/1942/26395 | DOI: | 10.1016/j.procs.2017.05.009 | ISI #: | 000404959000104 | Rights: | © 2017 The Authors. Published by Elsevier B.V. Peer-review under responsibility of the scientific committee of the International Conference on Computational Science | Category: | C1 | Type: | Proceedings Paper | Validations: | ecoom 2018 |
Appears in Collections: | Research publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Haber.pdf | Published version | 535.69 kB | Adobe PDF | View/Open |
SCOPUSTM
Citations
7
checked on Sep 3, 2020
WEB OF SCIENCETM
Citations
10
checked on Oct 12, 2024
Page view(s)
58
checked on Sep 7, 2022
Download(s)
220
checked on Sep 7, 2022
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.