English

A Primal-Dual Algorithmic Framework for Constrained Convex Minimization

Optimization and Control 2015-03-04 v2 Machine Learning

Abstract

We present a primal-dual algorithmic framework to obtain approximate solutions to a prototypical constrained convex optimization problem, and rigorously characterize how common structural assumptions affect the numerical efficiency. Our main analysis technique provides a fresh perspective on Nesterov's excessive gap technique in a structured fashion and unifies it with smoothing and primal-dual methods. For instance, through the choices of a dual smoothing strategy and a center point, our framework subsumes decomposition algorithms, augmented Lagrangian as well as the alternating direction method-of-multipliers methods as its special cases, and provides optimal convergence rates on the primal objective residual as well as the primal feasibility gap of the iterates for all.

Keywords

Cite

@article{arxiv.1406.5403,
  title  = {A Primal-Dual Algorithmic Framework for Constrained Convex Minimization},
  author = {Quoc Tran-Dinh and Volkan Cevher},
  journal= {arXiv preprint arXiv:1406.5403},
  year   = {2015}
}

Comments

This paper consists of 54 pages with 7 tables and 12 figures

R2 v1 2026-06-22T04:43:21.758Z