Conference proceeding
Online Nonnegative Matrix Factorization with General Divergences
ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 54, Vol.54, pp.37-45
Proceedings of Machine Learning Research
01/01/2017
Abstract
We develop a unified and systematic framework for performing online nonnegative matrix factorization under a wide variety of important divergences. The online nature of our algorithms makes them particularly amenable to large-scale data. We prove that the sequence of learned dictionaries converges almost surely to the set of critical points of the expected loss function. Experimental results demonstrate the computational efficiency and outstanding performances of our algorithms on several real-life applications, including topic modeling, document clustering and foreground-background separation.
Details
- Title: Subtitle
- Online Nonnegative Matrix Factorization with General Divergences
- Creators
- Renbo Zhao - Natl Univ Singapore, Singapore, SingaporeVincent Y. F. Tan - Natl Univ Singapore, Singapore, SingaporeHuan Xu - Georgia Institute of Technology
- Contributors
- Aarti Singh (Editor)Jerry Zhu (Editor)
- Resource Type
- Conference proceeding
- Publication Details
- ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 54, Vol.54, pp.37-45
- Publisher
- Microtome Publishing
- Series
- Proceedings of Machine Learning Research
- ISSN
- 2640-3498
- Number of pages
- 9
- Language
- English
- Date published
- 01/01/2017
- Academic Unit
- Business Analytics
- Record Identifier
- 9984446405602771
Metrics
3 Record Views