학술논문
MEGS: A Penalty for Mutually Exclusive Group Sparsity
Document Type
Periodical
Author
Source
IEEE Open Journal of Signal Processing IEEE Open J. Signal Process. Signal Processing, IEEE Open Journal of. 4:275-283 2023
Subject
Language
ISSN
2644-1322
Abstract
Penalty functions or regularization terms that promote structured solutions to optimization problems are of great interest in many fields. We introduce MEGS, a nonconvex structured sparsity penalty that promotes mutual exclusivity between components in solutions to optimization problems. This enforces, or promotes, 1-sparsity within arbitrary overlapping groups in a vector. The mutual exclusivity structure is represented by a matrix ${\bf {S}}$. We discuss the design of ${\bf {S}}$ from engineering principles and show example use cases including the modeling of occlusions in 3D imaging and a total variation variant with uses in image restoration. We also demonstrate synergy between MEGS and other regularizers and propose an algorithm to efficiently solve problems regularized or constrained by MEGS.