English

Global Convergence of Model Function Based Bregman Proximal Minimization Algorithms

Optimization and Control 2020-12-25 v1 Computer Vision and Pattern Recognition Machine Learning Numerical Analysis Numerical Analysis

Abstract

Lipschitz continuity of the gradient mapping of a continuously differentiable function plays a crucial role in designing various optimization algorithms. However, many functions arising in practical applications such as low rank matrix factorization or deep neural network problems do not have a Lipschitz continuous gradient. This led to the development of a generalized notion known as the LL-smad property, which is based on generalized proximity measures called Bregman distances. However, the LL-smad property cannot handle nonsmooth functions, for example, simple nonsmooth functions like \absx41\abs{x^4-1} and also many practical composite problems are out of scope. We fix this issue by proposing the MAP property, which generalizes the LL-smad property and is also valid for a large class of nonconvex nonsmooth composite problems. Based on the proposed MAP property, we propose a globally convergent algorithm called Model BPG, that unifies several existing algorithms. The convergence analysis is based on a new Lyapunov function. We also numerically illustrate the superior performance of Model BPG on standard phase retrieval problems, robust phase retrieval problems, and Poisson linear inverse problems, when compared to a state of the art optimization method that is valid for generic nonconvex nonsmooth optimization problems.

Keywords

Cite

@article{arxiv.2012.13161,
  title  = {Global Convergence of Model Function Based Bregman Proximal Minimization Algorithms},
  author = {Mahesh Chandra Mukkamala and Jalal Fadili and Peter Ochs},
  journal= {arXiv preprint arXiv:2012.13161},
  year   = {2020}
}

Comments

44 pages, 22 figures

R2 v1 2026-06-23T21:21:51.751Z