English

An $\widetilde{\mathcal{O}}(m/\varepsilon^{3.5})$-Cost Algorithm for Semidefinite Programs with Diagonal Constraints

Data Structures and Algorithms 2021-12-07 v2 Optimization and Control

Abstract

We study semidefinite programs with diagonal constraints. This problem class appears in combinatorial optimization and has a wide range of engineering applications such as in circuit design, channel assignment in wireless networks, phase recovery, covariance matrix estimation, and low-order controller design. In this paper, we give an algorithm to solve this problem to ε\varepsilon-accuracy, with a run time of O~(m/ε3.5)\widetilde{\mathcal{O}}(m/\varepsilon^{3.5}), where mm is the number of non-zero entries in the cost matrix. We improve upon the previous best run time of O~(m/ε4.5)\widetilde{\mathcal{O}}(m/\varepsilon^{4.5}) by Arora and Kale. As a corollary of our result, given an instance of the Max-Cut problem with nn vertices and mnm \gg n edges, our algorithm when applied to the standard SDP relaxation of Max-Cut returns a (1ε)αGW(1 - \varepsilon)\alpha_{GW} cut in time O~(m/ε3.5)\widetilde{\mathcal{O}}(m/\varepsilon^{3.5}), where αGW0.878567\alpha_{GW} \approx 0.878567 is the Goemans-Williamson approximation ratio. We obtain this run time via the stochastic variance reduction framework applied to the Arora-Kale algorithm, by constructing a constant-accuracy estimator to maintain the primal iterates.

Keywords

Cite

@article{arxiv.1903.01859,
  title  = {An $\widetilde{\mathcal{O}}(m/\varepsilon^{3.5})$-Cost Algorithm for Semidefinite Programs with Diagonal Constraints},
  author = {Yin Tat Lee and Swati Padmanabhan},
  journal= {arXiv preprint arXiv:1903.01859},
  year   = {2021}
}

Comments

Improved explanations (propagating changes from the conference version)

R2 v1 2026-06-23T07:58:44.934Z