A smoothing Newton method for second-order cone optimization based on a new smoothing function

Jingyong Tang

doi:10.1016/j.amc.2011.06.015

A smoothing Newton method for second-order cone optimization based on a new smoothing function

2011-10-01

20

PRE

AI

汤

汤京永 (Jingyong Tang) *

DOI：10.1016/j.amc.2011.06.015

Original

Original request for help

Save

Abstract

En 中文

A new smoothing function is given in this paper by smoothing the symmetric perturbed Fischer-Burmeister function. Based on this new smoothing function, we present a smoothing Newton method for solving the second-order cone optimization (SOCO). The method solves only one linear system of equations and performs only one line search at each iteration. Without requiring strict complementarity assumption at the SOCO solution, the proposed algorithm is shown to be globally and locally quadratically convergent. Numerical results demonstrate that our algorithm is promising and comparable to interior-point methods. (C) 2011 Elsevier Inc. All rights reserved.

Keywords:

Second-order cone optimization

Smoothing Newton method

Global convergence

Quadratic convergence

A smoothing Newton method for second-order cone optimization based on a new smoothing function

Abstract

Journal

Organization