Statistics > Computation
[Submitted on 9 Feb 2015 (v1), revised 15 May 2016 (this version, v4), latest version 5 Oct 2016 (v6)]
Title:Projected Nesterov's Proximal-Gradient Algorithm for Sparse Signal Reconstruction with a Convex Constraint
View PDFAbstract:We develop a projected Nesterov's proximal-gradient (PNPG) approach for sparse signal reconstruction that combines adaptive step size with Nesterov's momentum acceleration. The objective function that we wish to minimize is the sum of a convex differentiable data-fidelity (negative log-likelihood (NLL)) term and a convex regularization term. We apply sparse signal regularization where the signal belongs to a closed convex set within the closure of the domain of the NLL; the convex-set constraint facilitates flexible NLL domains and accurate signal recovery. Signal sparsity is imposed using the $\ell_1$-norm penalty on the signal's linear transform coefficients or gradient map, respectively. The PNPG approach employs projected Nesterov's acceleration step with restart and an inner iteration to compute the proximal mapping. We propose an adaptive step-size selection scheme to obtain a good local majorizing function of the NLL and reduce the time spent backtracking. Thanks to step-size adaptation, PNPG does not require Lipschitz continuity of the gradient of the NLL. We present an integrated derivation of the momentum acceleration and its $\mathcal{O}(k^{-2})$ convergence-rate and iterate convergence proofs, which account for adaptive step-size selection, inexactness of the iterative proximal mapping, and the convex-set constraint. The tuning of PNPG is largely application-independent. Tomographic and compressed-sensing reconstruction experiments with Poisson generalized linear and Gaussian linear measurement models demonstrate the performance of the proposed approach.
Submission history
From: Renliang Gu [view email][v1] Mon, 9 Feb 2015 19:19:46 UTC (1,798 KB)
[v2] Sat, 21 Feb 2015 21:25:37 UTC (1,759 KB)
[v3] Tue, 31 Mar 2015 16:25:34 UTC (1,752 KB)
[v4] Sun, 15 May 2016 19:41:15 UTC (481 KB)
[v5] Wed, 25 May 2016 21:42:27 UTC (481 KB)
[v6] Wed, 5 Oct 2016 08:04:30 UTC (546 KB)
Current browse context:
stat.CO
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.