Bayesian learning for robust principal component analysis
Authors:
- 1 August 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 2361-2365
Abstract
We develop a Bayesian learning method for robust principal component analysis where the main task is to estimate a low-rank matrix from noisy and outlier contaminated measurements. To promote low-rank, we use a structured Gaussian prior that induces correlations among column vectors as well as row vectors of the matrix under estimation. In our method, the noise and outliers are modeled by a combined noise model. The method is evaluated and compared to other methods using synthetic data as well as data from the MovieLens 100Kdataset. Comparisons show that the method empirically provides a significant performance improvement over existing methods.Keywords
This publication has 11 references indexed in Scilit:
- Relevance singular vector machine for low-rank matrix sensing2014
- Recovery of Low-Rank Plus Compressed Sparse Matrices With Application to Unveiling Traffic AnomaliesIEEE Transactions on Information Theory, 2013
- A unified approach to salient object detection via low rank matrix recovery2012
- Sparse Bayesian Methods for Low-Rank Matrix EstimationIEEE Transactions on Signal Processing, 2012
- Latent Variable Bayesian Models for Promoting SparsityIEEE Transactions on Information Theory, 2011
- Bayesian Robust Principal Component AnalysisIEEE Transactions on Image Processing, 2011
- Robust principal component analysis?Journal of the ACM, 2011
- Rank-Sparsity Incoherence for Matrix DecompositionSIAM Journal on Optimization, 2011
- Robust RVM regression using sparse outlier model2010
- Sparse Bayesian Learning for Basis SelectionIEEE Transactions on Signal Processing, 2004