English

Riemannian Smoothing Gradient Type Algorithms]{Riemannian Smoothing Gradient Type Algorithms for Nonsmooth Optimization Problem on Compact Riemannian Submanifold Embedded in Euclidean Space

Optimization and Control 2023-10-31 v3

Abstract

In this paper, we introduce the notion of generalized ϵ\epsilon-stationarity for a class of nonconvex and nonsmooth composite minimization problems on compact Riemannian submanifold embedded in Euclidean space. To find a generalized ϵ\epsilon-stationarity point, we develop a family of Riemannian gradient-type methods based on the Moreau envelope technique with a decreasing sequence of smoothing parameters, namely Riemannian smoothing gradient and Riemannian smoothing stochastic gradient methods. We prove that the Riemannian smoothing gradient method has the iteration complexity of O(ϵ3)\mathcal{O}(\epsilon^{-3}) for driving a generalized ϵ\epsilon-stationary point. To our knowledge, this is the best-known iteration complexity result for the nonconvex and nonsmooth composite problem on manifolds. For the Riemannian smoothing stochastic gradient method, one can achieve the iteration complexity of O(ϵ5)\mathcal{O}(\epsilon^{-5}) for driving a generalized ϵ\epsilon-stationary point. Numerical experiments are conducted to validate the superiority of our algorithms.

Keywords

Cite

@article{arxiv.2212.03526,
  title  = {Riemannian Smoothing Gradient Type Algorithms]{Riemannian Smoothing Gradient Type Algorithms for Nonsmooth Optimization Problem on Compact Riemannian Submanifold Embedded in Euclidean Space},
  author = {Zheng Peng and Weihe Wu and Jiang Hu and Kangkang Deng},
  journal= {arXiv preprint arXiv:2212.03526},
  year   = {2023}
}
R2 v1 2026-06-28T07:24:33.459Z