English

Delay-agnostic Asynchronous Coordinate Update Algorithm

Optimization and Control 2023-05-16 v1

Abstract

We propose a delay-agnostic asynchronous coordinate update algorithm (DEGAS) for computing operator fixed points, with applications to asynchronous optimization. DEGAS includes novel asynchronous variants of ADMM and block-coordinate descent as special cases. We prove that DEGAS converges under both bounded and unbounded delays under delay-free parameter conditions. We also validate by theory and experiments that DEGAS adapts well to the actual delays. The effectiveness of DEGAS is demonstrated by numerical experiments on classification problems.

Keywords

Cite

@article{arxiv.2305.08535,
  title  = {Delay-agnostic Asynchronous Coordinate Update Algorithm},
  author = {Xuyang Wu and Changxin Liu and Sindri Magnusson and Mikael Johansson},
  journal= {arXiv preprint arXiv:2305.08535},
  year   = {2023}
}

Comments

25 pages