Filters








76,065 Hits in 2.9 sec

A MAP approach for ℓ_q-norm regularized sparse parameter estimation using the EM algorithm [article]

Rodrigo Carvajal, Juan C. Agüero, Boris I. Godoy, Dimitrios Katselis
2015 arXiv   pre-print
In this paper, Bayesian parameter estimation through the consideration of the Maximum A Posteriori (MAP) criterion is revisited under the prism of the Expectation-Maximization (EM) algorithm.  ...  By incorporating a sparsity-promoting penalty term in the cost function of the estimation problem through the use of an appropriate prior distribution, we show how the EM algorithm can be used to efficiently  ...  In the next section, we will see how we can express the log-prior so that the EM algorithm can be used to tackle the q −norm regularized ML estimation problem.  ... 
arXiv:1508.01071v1 fatcat:5xpls7sqjjejpolbesxrb4oqwe

Design of a GF(64)-LDPC Decoder Based on the EMS Algorithm

Emmanuel Boutillon, Laura Conde-Canencia, Ali Al Ghouwayel
2013 IEEE Transactions on Circuits and Systems Part 1: Regular Papers  
The implemented decoder presents performance at less than 0.7 dB from the Belief Propagation algorithm for different code lengths and rates.  ...  This paper presents the architecture, performance and implementation results of a serial GF(64)-LDPC decoder based on a reduced-complexity version of the Extended Min-Sum algorithm.  ...  Min-Sum algorithm for NB-LDPC decoding The EMS algorithm [15] is an extension of the Min-Sum ( [39] [40] ) algorithm from binary to NB LDPC codes.  ... 
doi:10.1109/tcsi.2013.2279186 fatcat:tjdhkbbewngj3gjhe55a7utdca

Page 869 of Journal of Climate Vol. 14, Issue 5 [page]

2001 Journal of Climate  
| MARCH 2001 of the values is missing, the underestimation of the var- iances will be greater.* The test of the regularized EM algorithm demon- strates that the algorithm is applicable to typical sets  ...  The proofs that the conventional EM algorithm leads to consistent and unbiased estimators and converges mono- tonically (Little and Rubin 1987, chapter 7) are not transferable to the regularized EM algorithm  ... 

Comment on "Spatio-temporal filling of missing points in geophysical data sets" by D. Kondrashov and M. Ghil, Nonlin. Processes Geophys., 13, 151–159, 2006

T. Schneider
2007 Nonlinear Processes in Geophysics  
KG's method is just as parametric as the regularized EM algorithm.  ...  they reduce, in the limit of no regularization, to the EM algorithm for Gaussian data.  ...  they reduce, in the limit of no regularization, to the EM algorithm for Gaussian data.  ... 
doi:10.5194/npg-14-1-2007 fatcat:4a6aqmo3xbe2nkixcswsm4bpkq

Analysis of Incomplete Climate Data: Estimation of Mean Values and Covariance Matrices and Imputation of Missing Values

Tapio Schneider
2001 Journal of Climate  
values, is taken as the point of departure for the development of a regularized EM algorithm.  ...  In contrast to the conventional EM algorithm, the regularized EM algorithm is applicable to sets of climate data, in which the number of variables typically exceeds the sample size.  ...  I thank Keith Dixon for providing the ensemble of climate simulations; Isaac Held, John Lanzante, Michael Mann, and Arnold Neumaier for comments on drafts of this paper; and Heidi Swanson for editing the  ... 
doi:10.1175/1520-0442(2001)014<0853:aoicde>2.0.co;2 fatcat:jmlefbxsobd65f2gnovw5befju

Regularized online Mixture of Gaussians for background subtraction

Hongbin Wang, Paul Miller
2011 2011 8th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)  
Although the algorithm shows good empirical performance, it lacks theoretical justification.  ...  It is shown empirically that l 1 regularized online MoG converge faster than the original online MoG .  ...  Acknowledgement This research work is sponsored by the EPSRC projects EP/E028640/1 and EP/G034303/1.  ... 
doi:10.1109/avss.2011.6027331 dblp:conf/avss/WangM11 fatcat:cqyhizp4rzamfaueak4krqbdg4

EM-Type Algorithms for Image Reconstruction with Background Emission and Poisson Noise [chapter]

Ming Yan
2011 Lecture Notes in Computer Science  
In order to overcome the contrast reduction introduced by some regularizations, we suggested EM-Type algorithms with Bregman iteration by applying a sequence of modified EM-Type algorithms.  ...  This method is separated into two steps: EM and regularization.  ...  The regularization we used is total variation (TV) regularization, and we present some deblurring results with the proposed EM-TV algorithm and the Bregman version of it.  ... 
doi:10.1007/978-3-642-24028-7_4 fatcat:xwapcrlmazag7oewkasrcrr6bq

Accurate EM-TV algorithm in PET with low SNR

Alex Sawatzky, Christoph Brune, Frank Wubbeling, Thomas Kosters, Klaus Schafers, Martin Burger
2008 2008 IEEE Nuclear Science Symposium Conference Record  
In these cases standard reconstruction methods (OSEM, EM, filtered backprojection) deliver unsatisfactory and noisy results.  ...  The general approach can also be used for other specific goals allowing to incorporate a-priori information about the solution with Poisson distributed data.  ...  In the absence of regularization (α = 0) the EM-algorithm (cf. [2] ) has become a standard scheme, which is however difficult to be generalized to regularized cases.  ... 
doi:10.1109/nssmic.2008.4774392 fatcat:yb2vkjxmlffjxlmvsdekv6t65y

A globally convergent regularized ordered-subset EM algorithm for list-mode reconstruction

P. Khurd, Ing-Tsung Hsiao, A. Rangarajan, G. Gindi
2004 IEEE Transactions on Nuclear Science  
In this work, we present a globally convergent and regularized orderedsubset algorithm for LM reconstruction. Our algorithm is derived using an incremental EM approach.  ...  We investigated the speed-up of our LM OS algorithm (vs. a non-OS version) for a SPECT simulation, and found that the speed-up was somewhat less than that enjoyed by other OS-type algorithms.  ...  LM-EM-MAP: ADDITION OF REGULARIZATION Though not done in [9] , one could add regularization to this EM algorithm to derive a non-OS list mode MAP algorithm.  ... 
doi:10.1109/tns.2004.829780 fatcat:hwishpsmr5b4ta25i77dzg7oai

AN ACCELERATED ALGORITHM FOR DENSITY ESTIMATION IN LARGE DATABASES USING GAUSSIAN MIXTURES

Alvaro Soto, Felipe Zavala, Anita Araneda
2007 Cybernetics and systems  
In this paper we introduce an EM-based algorithm, that solves the scalability problem.  ...  The classical estimation approach for GMMs corresponds to the iterative algorithm of Expectation Maximization.  ...  We compare AGMM-EM to the results of popular existing algorithms, as are regular EM, KD-Clust (?), and FASTEM (?).  ... 
doi:10.1080/01969720601138928 fatcat:yheisqjfdffu5ly4ggecusmr6m

A method of attenuation map and emission activity reconstruction from emission data

V.Y. Panin, G.L. Zeng, G.T. Gullberg
2001 IEEE Transactions on Nuclear Science  
The algorithm is based on the quasilinearized attenuated Radon transform.  ...  At each iteration of the algorithm, emission activity distribution is reconstructed using the attenuation map from the previous iteration.  ...  ACKNOWLEDGMENTS We thank Sean Webb for editing the manuscript.  ... 
doi:10.1109/23.910843 fatcat:ooyefumsljcphfugxopkadwt4a

Unambiguity Regularization for Unsupervised Learning of Probabilistic Grammars

Kewei Tu, Vasant G. Honavar
2012 Conference on Empirical Methods in Natural Language Processing  
The resulting family of algorithms includes the expectation-maximization algorithm (EM) and its variant, Viterbi EM, as well as a so-called softmax-EM algorithm.  ...  The softmax-EM algorithm can be implemented with a simple and computationally efficient extension to standard EM.  ...  The work of Vasant Honavar was supported by the National Science Foundation, while working at the Foundation.  ... 
dblp:conf/emnlp/TuH12 fatcat:klcucroicbamxgj2hbkkymn33m

Bregman-EM-TV Methods with Application to Optical Nanoscopy [chapter]

Christoph Brune, Alex Sawatzky, Martin Burger
2009 Lecture Notes in Computer Science  
Starting from a statistical modeling in terms of a MAP likelihood estimation we combine the iterative EM algorithm with TV regularization techniques to make an efficient use of a-priori information.  ...  In these applications standard reconstruction methods (EM, filtered backprojection) deliver unsatisfactory and noisy results.  ...  This work has been supported by the German Federal Ministry of Education and Research through the project INVERS.  ... 
doi:10.1007/978-3-642-02256-2_20 fatcat:yix4gr7n5bd3rdlsz7lnb4nimm

Reconstruction of X-ray Fluorescence Computed Tomography from Sparse-View Projections via L1-norm Regularized EM Algorithm

Junwei Shi, Daiki Hara, Wensi Tao, Nesrin Dogan, Alan Pollack, John Chetley Ford
2020 IEEE Access  
As a kind of Green's one-step-late algorithm, the proposed L1-EM algorithm relies on an appropriate regularization parameter to achieve the expected results.  ...  The implementation of this algorithm takes into account the simplicity of the emission-EM-look-like algorithms and the robustness of L1-norm regularization against artifacts.  ... 
doi:10.1109/access.2020.3039927 fatcat:2ua57tqa6vea7gz2mwibqegsua

General convergent expectation maximization (EM)-type algorithms for image reconstruction

Ming Yan, Alex A. T. Bui, Jason Cong, Luminita Vese
2013 Inverse Problems and Imaging  
The EM-Type algorithms are performed using iteratively EM (or SART for weighted Gaussian noise) and regularization in the image domain.  ...  To show the efficiency of EM-Type algorithms, the application in computerized tomography reconstruction is chosen.  ...  The EM-Type algorithms are performed using iteratively EM (or SART for weighted Gaussian noise) and regularization in the image domain.  ... 
doi:10.3934/ipi.2013.7.1007 fatcat:2wrojk4onvb3tdfdhk773oi3qy
« Previous Showing results 1 — 15 out of 76,065 results