Sparse Estimation with Generalized Beta Mixture and the Horseshoe Prior

11/10/2014
by   Zahra Sabetsarvestani, et al.
0

In this paper, the use of the Generalized Beta Mixture (GBM) and Horseshoe distributions as priors in the Bayesian Compressive Sensing framework is proposed. The distributions are considered in a two-layer hierarchical model, making the corresponding inference problem amenable to Expectation Maximization (EM). We present an explicit, algebraic EM-update rule for the models, yielding two fast and experimentally validated algorithms for signal recovery. Experimental results show that our algorithms outperform state-of-the-art methods on a wide range of sparsity levels and amplitudes in terms of reconstruction accuracy, convergence rate and sparsity. The largest improvement can be observed for sparse signals with high amplitudes.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/23/2017

An Expectation Maximization Framework for Preferential Attachment Models

In this paper we develop an Expectation Maximization(EM) algorithm to es...
research
01/22/2016

Rectified Gaussian Scale Mixtures and the Sparse Non-Negative Least Squares Problem

In this paper, we develop a Bayesian evidence maximization framework to ...
research
11/06/2017

Simultaneous Block-Sparse Signal Recovery Using Pattern-Coupled Sparse Bayesian Learning

In this paper, we consider the block-sparse signals recovery problem in ...
research
08/24/2017

Bayesian Compressive Sensing Using Normal Product Priors

In this paper, we introduce a new sparsity-promoting prior, namely, the ...
research
07/14/2017

Hierarchical EM algorithm for estimating the parameters of Mixture of Bivariate Generalized Exponential distributions

This paper provides a mixture modeling framework using the bivariate gen...
research
10/17/2018

Generalized Earthquake Frequency-Magnitude Distribution Described by Asymmetric Laplace Mixture Modelling

The complete part of the earthquake frequency-magnitude distribution (FM...
research
04/19/2015

Exploring Bayesian Models for Multi-level Clustering of Hierarchically Grouped Sequential Data

A wide range of Bayesian models have been proposed for data that is divi...

Please sign up or login with your details

Forgot password? Click here to reset