References & Citations
Statistics > Computation
Title: Minimizing Sum of Truncated Convex Functions and Its Applications
(Submitted on 31 Jul 2016 (v1), last revised 27 Jun 2017 (this version, v3))
Abstract: In this paper, we study a class of problems where the sum of truncated convex functions is minimized. In statistical applications, they are commonly encountered when $\ell_0$-penalized models are fitted and usually lead to NP-Hard non-convex optimization problems. In this paper, we propose a general algorithm for the global minimizer in low-dimensional settings. We also extend the algorithm to high-dimensional settings, where an approximate solution can be found efficiently. We introduce several applications where the sum of truncated convex functions is used, compare our proposed algorithm with other existing algorithms in simulation studies, and show its utility in edge-preserving image restoration on real data.
Submission history
From: Hui Jiang [view email][v1] Sun, 31 Jul 2016 16:14:13 GMT (342kb,D)
[v2] Sat, 22 Oct 2016 01:59:24 GMT (1021kb,D)
[v3] Tue, 27 Jun 2017 03:05:49 GMT (5402kb,D)
Link back to: arXiv, form interface, contact.