Biomedical Imaging GroupSTI
English only   BIG > Publications > Learned Regularizers

 Home Page
 News & Events
 Tutorials and Reviews
 Download Algorithms

 All BibTeX References

Learning Convex Regularizers for Optimal Bayesian Denoising

H.Q. Nguyen, E. Bostan, M. Unser

IEEE Transactions on Signal Processing, vol. 66, no. 4, pp. 1093-1105, February 15, 2018.

We propose a data-driven algorithm for the Bayesian estimation of stochastic processes from noisy observations. The primary statistical properties of the sought signal are specified by the penalty function (i.e., negative logarithm of the prior probability density function). Our alternating direction method of multipliers (ADMM) based approach translates the estimation task into successive applications of the proximal mapping of the penalty function. Capitalizing on this direct link, we define the proximal operator as a parametric spline curve and optimize the spline coefficients by minimizing the average reconstruction error for a given training set. The key aspects of our learning method are that the associated penalty function is constrained to be convex and the convergence of the ADMM iterations is proven. As a result of these theoretical guarantees, adaptation of the proposed framework to different levels of measurement noise is extremely simple and does not require any retraining. We apply our method to estimation of both sparse and nonsparse models of Lévy processes for which the minimum mean square error (MMSE) estimators are available. We carry out a single training session for a fixed level of noise and perform comparisons at various signal-to-noise ratio values. Simulations illustrate that the performance of our algorithm are practically identical to the one of the MMSE estimator irrespective of the noise power.

AUTHOR="Nguyen, H.Q. and Bostan, E. and Unser, M.",
TITLE="Learning Convex Regularizers for Optimal {B}ayesian Denoising",
JOURNAL="{IEEE} Transactions on Signal Processing",
month="February 15,",

© 2018 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from IEEE.
This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be reposted without the explicit permission of the copyright holder.