eduzhai > Applied Sciences > Engineering >

SWAGGER Sparsity Within and Across Groups for General Estimation and Recovery

  • Save

... pages left unread,continue reading

Document pages: 8 pages

Abstract: Penalty functions or regularization terms that promote structured solutionsto optimization problems are of great interest in many fields. Proposed in thiswork is a nonconvex structured sparsity penalty that promotes one-sparsitywithin arbitrary overlapping groups in a vector. This allows one to enforcemutual exclusivity between components within solutions to optimizationproblems. We show multiple example use cases (including a total variationvariant), demonstrate synergy between it and other regularizers, and propose analgorithm to efficiently solve problems regularized or constrained by theproposed penalty.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×