Skip to main content
Cornell University
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > eess > arXiv:2003.04798v1

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Electrical Engineering and Systems Science > Signal Processing

arXiv:2003.04798v1 (eess)
[Submitted on 10 Mar 2020 (this version), latest version 26 Oct 2020 (v3)]

Title:Convergence Guarantees for Non-Convex Optimisation with Cauchy-Based Penalties

Authors:Oktay Karakus, Perla Mayo, Alin Achim
View a PDF of the paper titled Convergence Guarantees for Non-Convex Optimisation with Cauchy-Based Penalties, by Oktay Karakus and 2 other authors
View PDF
Abstract:In this paper, we propose a convex proximal splitting methodology with a non-convex penalty function based on the heavy-tailed Cauchy distribution. We first suggest a closed-form expression for calculating the proximal operator of the Cauchy prior, which then makes it applicable in generic proximal splitting algorithms. We further derive the required condition for minimisation problems with the Cauchy based penalty function that guarantees the convergence to the global minimum even though it is non-convex. Setting the system parameters by satisfying the proposed condition keeps the overall cost function convex, which is minimised via the forward-backward (FB) algorithm. The Cauchy regularisation based proposed method is evaluated in solving two generic signal processing examples, which are a 1D signal denoising in the frequency domain and two 2D image reconstruction studies of \textit{de-blurring} and \textit{denoising}. We experimentally verify the proposed convexity conditions for various cases, and show the effectiveness of the proposed Cauchy based non-convex penalty function over the state-of-the-art penalty functions of $L_1$ and total variation ($TV$) norms.
Comments: 17 pages, 7 figures
Subjects: Signal Processing (eess.SP); Image and Video Processing (eess.IV)
Cite as: arXiv:2003.04798 [eess.SP]
  (or arXiv:2003.04798v1 [eess.SP] for this version)
  https://doi.org/10.48550/arXiv.2003.04798
arXiv-issued DOI via DataCite

Submission history

From: Oktay Karakuş Dr [view email]
[v1] Tue, 10 Mar 2020 15:23:29 UTC (723 KB)
[v2] Sat, 28 Mar 2020 18:43:10 UTC (487 KB)
[v3] Mon, 26 Oct 2020 12:30:46 UTC (1,479 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Convergence Guarantees for Non-Convex Optimisation with Cauchy-Based Penalties, by Oktay Karakus and 2 other authors
  • View PDF
  • Other Formats
view license
Current browse context:
eess.SP
< prev   |   next >
new | recent | 2020-03
Change to browse by:
eess
eess.IV

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar
a export BibTeX citation Loading...

BibTeX formatted citation

×
Data provided by:

Bookmark

BibSonomy logo Reddit logo

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
  • Author
  • Venue
  • Institution
  • Topic

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.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)
  • About
  • Help
  • contact arXivClick here to contact arXiv Contact
  • subscribe to arXiv mailingsClick here to subscribe Subscribe
  • Copyright
  • Privacy Policy
  • Web Accessibility Assistance
  • arXiv Operational Status
    Get status notifications via email or slack