English

On the Curved Geometry of Accelerated Optimization

Machine Learning 2019-11-21 v2 Optimization and Control Machine Learning

Abstract

In this work we propose a differential geometric motivation for Nesterov's accelerated gradient method (AGM) for strongly-convex problems. By considering the optimization procedure as occurring on a Riemannian manifold with a natural structure, The AGM method can be seen as the proximal point method applied in this curved space. This viewpoint can also be extended to the continuous time case, where the accelerated gradient method arises from the natural block-implicit Euler discretization of an ODE on the manifold. We provide an analysis of the convergence rate of this ODE for quadratic objectives.

Keywords

Cite

@article{arxiv.1812.04634,
  title  = {On the Curved Geometry of Accelerated Optimization},
  author = {Aaron Defazio},
  journal= {arXiv preprint arXiv:1812.04634},
  year   = {2019}
}

Comments

NeurIPS 2019 Accepted paper

R2 v1 2026-06-23T06:39:27.595Z