An EM algorithm for learning sparse and overcomplete representations

Mingjun Zhong; Huanwen Tang; Hongjun Chen; Yiyuan Tang

doi:10.1016/j.neucom.2003.12.002

An EM algorithm for learning sparse and overcomplete representations

Mingjun Zhong, Huanwen Tang, Hongjun Chen, Yiyuan Tang

School of Informatics

Research output: Contribution to journal › Article › peer-review

Abstract / Description of output

An expectation-maximization (EM) algorithm for learning sparse and overcomplete representations is presented in this paper. We show that the estimation of the conditional moments of the posterior distribution can be accomplished by maximum a posteriori estimation. The approximate conditional moments enable the development of an EM algorithm for learning the overcomplete basis vectors and inferring the most probable basis coefficients.

Original language	English
Pages (from-to)	469-476
Number of pages	8
Journal	Neurocomputing
Volume	57
DOIs	https://doi.org/10.1016/j.neucom.2003.12.002
Publication status	Published - Mar 2004

Keywords / Materials (for Non-textual outputs)

Overcomplete representations
EM algorithm
Maximum a posteriori

Access to Document

10.1016/j.neucom.2003.12.002

http://www.sciencedirect.com/science/article/pii/S092523120300540X

Cite this

@article{b0e80bf150b94f97a22417bb8d143527,

title = "An EM algorithm for learning sparse and overcomplete representations",

abstract = "An expectation-maximization (EM) algorithm for learning sparse and overcomplete representations is presented in this paper. We show that the estimation of the conditional moments of the posterior distribution can be accomplished by maximum a posteriori estimation. The approximate conditional moments enable the development of an EM algorithm for learning the overcomplete basis vectors and inferring the most probable basis coefficients.",

keywords = "Overcomplete representations, EM algorithm, Maximum a posteriori",

author = "Mingjun Zhong and Huanwen Tang and Hongjun Chen and Yiyuan Tang",

year = "2004",

month = mar,

doi = "10.1016/j.neucom.2003.12.002",

language = "English",

volume = "57",

pages = "469--476",

journal = "Neurocomputing",

issn = "0925-2312",

publisher = "Elsevier",

}

TY - JOUR

T1 - An EM algorithm for learning sparse and overcomplete representations

AU - Zhong, Mingjun

AU - Tang, Huanwen

AU - Chen, Hongjun

AU - Tang, Yiyuan

PY - 2004/3

Y1 - 2004/3

N2 - An expectation-maximization (EM) algorithm for learning sparse and overcomplete representations is presented in this paper. We show that the estimation of the conditional moments of the posterior distribution can be accomplished by maximum a posteriori estimation. The approximate conditional moments enable the development of an EM algorithm for learning the overcomplete basis vectors and inferring the most probable basis coefficients.

AB - An expectation-maximization (EM) algorithm for learning sparse and overcomplete representations is presented in this paper. We show that the estimation of the conditional moments of the posterior distribution can be accomplished by maximum a posteriori estimation. The approximate conditional moments enable the development of an EM algorithm for learning the overcomplete basis vectors and inferring the most probable basis coefficients.

KW - Overcomplete representations

KW - EM algorithm

KW - Maximum a posteriori

U2 - 10.1016/j.neucom.2003.12.002

DO - 10.1016/j.neucom.2003.12.002

M3 - Article

SN - 0925-2312

VL - 57

SP - 469

EP - 476

JO - Neurocomputing

JF - Neurocomputing

ER -

An EM algorithm for learning sparse and overcomplete representations

Abstract / Description of output

Keywords / Materials (for Non-textual outputs)

Access to Document

Fingerprint

Cite this