logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

Algorithm xxxx: HDSDP: Software for Semidefinite Programming

delete2025-02-28
delete0
PRE
AI
W
Wenzhi Gao
D
Dongdong Ge
Y
Yinyu Ye
DOI：10.1145/3721123delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
<jats:p> <jats:monospace>HDSDP</jats:monospace> is a numerical software solving semidefinite programming problems. The main framework of <jats:monospace>HDSDP</jats:monospace> resembles the dual-scaling interior point solver <jats:monospace>DSDP</jats:monospace> [Benson and Ye, 2008] and several new features, including a dual method based on the simplified homogeneous self-dual embedding, have been implemented. The embedding technique enhances the stability of the dual method , and several new heuristics and computational techniques are designed to accelerate its convergence. HDSDP aims to show how the dual-scaling algorithm benefits from the self-dual embedding, and it is developed in parallel to DSDP5.8. Numerical experiments over several classical benchmark datasets exhibit its robustness and efficiency, particularly its advantages on SDP instances featuring low-rank structure and sparsity. <jats:monospace>HDSDP</jats:monospace> is open-sourced under an MIT license and available at <jats:ext-link xmlns:xlink="http://www.w3.org/1999/xlink" ext-link-type="url" xlink:href="https://github.com/Gwzwpxz/HDSDP">https://github.com/Gwzwpxz/HDSDP</jats:ext-link> . </jats:p>
Keywords:
semidefinite programming
dual-scaling algorithm
homogeneous self-dual embedding
interior point method
low-rank structure

Journal

ACM Transactions on Mathematical Software cover
ACM Transactions on Mathematical Software
IF:
3.2
Papers:
33
Citations:
5.1K

Organization

No organization information available