Sharpening Sparse Regularizers via Smoothing
Non-convex sparsity-inducing penalties outperform their convex counterparts, but generally sacrifice the cost function convexity. As a middle ground, we propose the <italic>sharpening sparse regularizers</italic> (SSR) framework to design non-separable non-convex penalties that induce sp...
Main Authors: | Abdullah H. Al-Shabili, Yining Feng, Ivan Selesnick |
---|---|
Format: | Article |
Language: | English |
Published: |
IEEE
2021-01-01
|
Series: | IEEE Open Journal of Signal Processing |
Subjects: | |
Online Access: | https://ieeexplore.ieee.org/document/9512409/ |
Similar Items
-
An introduction to non-smooth convex analysis via multiplicative derivative
by: Ali Hakan Tor
Published: (2019-12-01) -
Convex matrix sparsity for demixing with an application to graphical model structure estimation
by: Vinyes, Marina
Published: (2018) -
Robust Fault Detection Based on l1 Regularization
by: Young-Man Kim
Published: (2021-04-01) -
Deblurring with Framelets in the Sparse Analysis Setting
by: Danniels, Travis
Published: (2013) -
Salient Object Detection Based on Unified Convex Surrogate for Non-Convex Schatten-<inline-formula> <tex-math notation="LaTeX">$p$ </tex-math></inline-formula> Norm
by: Min Li, et al.
Published: (2020-01-01)