Infinite Sparse Factor Analysis and Infinite Independent Components Analysis
作者： David Knowles , Zoubin Ghahramani , None
DOI: 10.1007/978-3-540-74494-8_48
关键词:
摘要: A nonparametric Bayesian extension of Independent Components Analysis (ICA) is proposed where observed data Y modelled as a linear superposition, G, potentially infinite number hidden sources, X. Whether given source active for specific point specified by an binary matrix, Z. The resulting sparse representation allows increased reduction compared to standard ICA. We define prior on Z using the Indian Buffet Process (IBP). describe four variants model, with Gaussian or Laplacian priors X and one two-parameter IBPs. demonstrate inference under these models Markov Chain Monte Carlo (MCMC) algorithm synthetic gene expression compare ICA algorithms.
springer.com
springerlink.com参考文章(10)
Sylvia. Richardson, Peter J. Green, On Bayesian Analysis of Mixtures with an Unknown Number of Components (with discussion) Journal of the Royal Statistical Society: Series B (Statistical Methodology). ,vol. 59, pp. 731- 792 ,(1997) , 10.1111/1467-9868.00095
Sylvia Richardson, Peter J. Green, On Bayesian Analysis of Mixtures with an Unknown Number of Components ,(1997)
Z Ghahramani, P Sollich, TL Griffiths, None, Rejoinder for "Bayesian Nonparametric Latent Feature Models" ,(2007)
Zoubin Ghahramani, Edward Meeds, Sam T. Roweis, Radford M. Neal, Modeling Dyadic Data with Binary Latent Factors neural information processing systems. ,vol. 19, pp. 977- 984 ,(2006)
Zoubin Ghahramani, Thomas L. Griffiths, Infinite latent feature models and the Indian buffet process neural information processing systems. ,vol. 18, pp. 475- 482 ,(2005)
Shun-ichi Amari, Andrzej Cichocki, Howard Hua Yang, A New Learning Algorithm for Blind Signal Separation neural information processing systems. ,vol. 8, pp. 757- 763 ,(1995)
A. Hyvarinen, Fast and robust fixed-point algorithms for independent component analysis IEEE Transactions on Neural Networks. ,vol. 10, pp. 626- 634 ,(1999) , 10.1109/72.761722
A.-M. Martoglio, J. W. Miskin, S. K. Smith, D. J. C. MacKay, A decomposition model to track gene expression signatures: preview on observer-independent classification of ovarian cancer. Bioinformatics. ,vol. 18, pp. 1617- 1624 ,(2002) , 10.1093/BIOINFORMATICS/18.12.1617
Scott Makeig, Terrence J. Sejnowski, Tzyy-Ping Jung, Anthony J. Bell, Independent Component Analysis of Electroencephalographic Data neural information processing systems. ,vol. 8, pp. 145- 151 ,(1995)
D. Heckerman, Julia Lasserre, J. O. Berger, M. West, Christopher Bishop, J. M. Bernardo, M. J. Bayarri, A. P. Dawid, A. F. M. Smith, Bayesian Statistics 8 Oxford University Press. ,(2007)