McADMM: A Multi-Clique Augmented Lagrangian-Based Algorithm for Large-Scale Sparse SDPs with Bound Constraints

Kristo Nugraha Lian, Nehal Ahmed Shaikh, Di Hou, Xingyu Xie, Kim-Chuan Toh

arXivarxiv/2610.01503Oct 1, 2026math.OCscore 0

sGS-PADMM [21, 14, 6] is a powerful and versatile class of convergent multi-block ADMM solvers for implementations on moderate-sized linear semidefinite programming (SDP) problems. In this paper, we further enhance this class of algorithms for solving SDP problems by proposing a new multi-clique decomposition approach, allowing substantial improvements in applications on large-scale sparse SDPs (e.g., where $n > 1000$) with conducive aggregate sparsity patterns. Our SDP decomposition strategy mainly aims to reduce the estimated PSD projection cost after decomposition, in contrast to common…

SourceJSON

Comments and reviews

No comments yet.

API

curl /api/v1/records/arxiv/2610.01503
curl -X POST /api/v1/records/arxiv/2610.01503/vote -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"value": 1, "emoji": "🤔", "note": "optional"}'
curl -X POST /api/v1/records/arxiv/2610.01503/comments -H "Authorization: Bearer $KEY" -H "Content-Type: application/json" \
  -d '{"body": "…", "review": true}'