Hierarchical Infinite Divisibility for Multiscale Shrinkage
Open Access
- 1 July 2014
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Signal Processing
- Vol. 62 (17) , 4363-4374
Abstract
A new shrinkage-based construction is developed for a compressible vector ${\mmb x}\in\BBR^{n}$ , for cases in which the components of ${\mmb x}$ are naturally associated with a tree structure. Important examples are when ${\mmb x}$ corresponds to the coefficients of a wavelet or block-DCT representation of data. The method we consider in detail, and for which numerical results are presented, is based on the gamma distribution. The gamma distribution is a heavy-tailed distribution that is infinitely divisible, and these characteristics are leveraged within the model. We further demonstrate that the general framework is appropriate for many other types of infinitely divisible heavy-tailed distributions. Bayesian inference is carried out by approximating the posterior with samples from an MCMC algorithm, as well as by constructing a variational approximation to the posterior. We also consider expectation-maximization (EM) for a MAP (point) solution. State-of-the-art results are manifested for compressive sensing and denoising applications, the latter with spiky (non-Gaussian) noise.
Keywords
Funding Information
- ARO
- DARPA
- DOE
- NGA
- ONR
This publication has 41 references indexed in Scilit:
- A Unified Formulation of Gaussian Versus Sparse Stochastic Processes—Part I: Continuous-Domain TheoryIEEE Transactions on Information Theory, 2014
- Sparse Stochastic Processes and Discretization of Linear Inverse ProblemsIEEE Transactions on Image Processing, 2013
- MMSE Estimation of Sparse Lévy ProcessesIEEE Transactions on Signal Processing, 2012
- Robust principal component analysis?Journal of the ACM, 2011
- The Bayesian LassoJournal of the American Statistical Association, 2008
- Fast bayesian matching pursuit2008
- Compressive Sensing [Lecture Notes]IEEE Signal Processing Magazine, 2007
- Isotropic polyharmonic B-splines: scaling functions and waveletsIEEE Transactions on Image Processing, 2005
- Lévy Processes and Stochastic Calculus2004
- Wavelet-based statistical signal processing using hidden Markov modelsIEEE Transactions on Signal Processing, 1998