English

Speeding up Stochastic Proximal Optimization in the High Hessian Dissimilarity Setting

Optimization and Control 2024-12-19 v1

Abstract

Stochastic proximal point methods have recently garnered renewed attention within the optimization community, primarily due to their desirable theoretical properties. Notably, these methods exhibit a convergence rate that is independent of the Lipschitz smoothness constants of the loss function, a feature often missing in the loss functions of modern ML applications. In this paper, we revisit the analysis of the Loopless Stochastic Variance Reduced Proximal Point Method (L-SVRP). Building on existing work, we establish a theoretical improvement in the convergence rate in scenarios characterized by high Hessian dissimilarity among the functions. Our concise analysis, which does not require smoothness assumptions, demonstrates a significant improvement in communication complexity compared to standard stochastic gradient descent.

Keywords

Cite

@article{arxiv.2412.13619,
  title  = {Speeding up Stochastic Proximal Optimization in the High Hessian Dissimilarity Setting},
  author = {Elnur Gasanov and Peter Richtárik},
  journal= {arXiv preprint arXiv:2412.13619},
  year   = {2024}
}
R2 v1 2026-06-28T20:40:05.984Z