Please use this identifier to cite or link to this item:
http://hdl.handle.net/1942/26257
Title: | Macau: scalable Bayesian factorization with high-dimensional side information using MCMC | Authors: | Simm, J. Arany, A. Zakeri, P. HABER, Tom Wegner, J. K. Chupakhin, V. Ceulemans, H. Moreau, Y. |
Issue Date: | 2017 | Publisher: | IEEE | Source: | Ueda, Naonori; Watanabe, Shinji; Matsui, Tomoko; Chien, Jen-Tzung; Larsen, Jan (Ed.). Proceedings of MLSP2017, IEEE, | Series/Report: | IEEE International Workshop on Machine Learning for Signal Processing | Abstract: | Bayesian matrix factorization is a method of choice for making predictions for large-scale incomplete matrices, due to availability of efficient Gibbs sampling schemes and its robustness to overfitting. In this paper, we consider factorization of large scale matrices with high-dimensional side information. However, sampling the link matrix for the side information with standard approaches costs O (F-3) time, where F is the dimensionality of the features. To overcome this limitation we, firstly, propose a prior for the link matrix whose strength is proportional to the scale of latent variables. Secondly, using this prior we derive an efficient sampler, with linear complexity in the number of non-zeros, O (N-nz), by leveraging Krylov subspace methods, such as block conjugate gradient, allowing us to handle million-dimensional side information. We demonstrate the effectiveness of our proposed method in drug-protein interaction prediction task. | Notes: | Simm, J (reprint author), Katholieke Univ Leuven, ESAT STADIUS, B-3001 Heverlee, Belgium, | Keywords: | matrix factorization; side information; high scale machine learning; MCMC | Document URI: | http://hdl.handle.net/1942/26257 | ISBN: | 9781509063413 | DOI: | 10.1109/MLSP.2017.8168143 | ISI #: | 000425458700038 | Rights: | Copyright ©2017 by the Institute of Electrical and Electronics Engineers, Inc. All rights reserved | Category: | C1 | Type: | Proceedings Paper | Validations: | ecoom 2019 |
Appears in Collections: | Research publications |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
macau.pdf Restricted Access | Published version | 177.88 kB | Adobe PDF | View/Open Request a copy |
SCOPUSTM
Citations
9
checked on Sep 3, 2020
WEB OF SCIENCETM
Citations
20
checked on Oct 12, 2024
Page view(s)
52
checked on Sep 5, 2022
Download(s)
44
checked on Sep 5, 2022
Google ScholarTM
Check
Altmetric
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.