VBMF source code

VBMF_SMILE is a simple MATLAB package for learning variational Bayesian matrix factorization (VBMF). It currently supports four different learning algorithms – VBMFbias_BCDfull [1], VBMFbias_GD [2], VBMbias_CD [3], and VBMFbiasSIDEINFO_CD [4].
Note that VBMFbias_CD and VBMFbiasSIDEINFO_CD are suitable for large scale dataset.

Please see Readme.txt for more detailed information

Category Software 0 / 0
License GPL v2
Producer Admin
Package Regdate Jun 24, 2015 11:52 Package Downloaded 57
Latest version 1.0 Download
Downloaded Count 57
Recent Updates Jun 24, 2015 11:54 0 / 0

Introduction

VBMF_SMILE is a simple MATLAB package for learning variational Bayesian matrix factorization (VBMF). It currently supports four different learning algorithms – VBMFbias_BCDfull [1], VBMFbias_GD [2], VBMbias_CD [3], and VBMFbiasSIDEINFO_CD [4].

Note that VBMFbias_CD and VBMFbiasSIDEINFO_CD are suitable for large scale dataset.

 

Quick Start

At first add the VBMF_SMILE folder to MATLAB Path.

Please run “demo.m”. It will load the MovieLens dataset, and then it sequentially run each algorithm.

 

How to Cite

If you find VBMF_SMILE helpful, please cite it as

 

Yong-Deok Kim and Seungjin Choi.
Scalable variational Bayesian matrix factorization with side information.
In Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics (AISTATS-2014), Reykjavik, Iceland, April 22-25, 2014.

 

Contact Information

For any questions and comments, please send your email to yongdeok.kim.mlg@gmail.com

 

References

[1] Y. J. Lim and Y. W. Teh.

Variational Bayesian approach to movie rating prediction.

In Proceedings of KDD Cup and Workshop, San Jose, CA, 2007.

 

[2] T. Raiko, A. Ilin, and J. Karhunen.

Principal component analysis for large scale problems with lots of missing values.

In Proceedings of the European Conference on Machine Learning (ECML), pages 691–698, Warsaw, Poland, 2007.

 

[3] Y. –D. Kim and Seunjin Choi.

Scalable Variational Bayesian Matrix Factorization.

In Proceedings of the First Workshop on Large-Scale Recommender Systems (LSRS-2013), Hong Kong, October 13, 2013.

 

[4] Yong-Deok Kim and Seungjin Choi.
Scalable variational Bayesian matrix factorization with side information.
In Proceedings of the Seventeenth International Conference on Artificial Intelligence and Statistics (AISTATS-2014), Reykjavik, Iceland, April 22-25, 2014.