Analysis of Sequential Quadratic Programming through the Lens of Riemannian Optimization

arXiv1805.08756MaRDI QIDQ6301951FDOQ6301951

Authors: Yu Bai, Song Mei 

Publication date: 22 May 2018

Abstract: We prove that a "first-order" Sequential Quadratic Programming (SQP) algorithm for equality constrained optimization has local linear convergence with rate $(1-1/kapp{a}_R{)}^k$, where $kapp{a}_R$ is the condition number of the Riemannian Hessian, and global convergence with rate $k^{-1/4}$. Our analysis builds on insights from Riemannian optimization -- we show that the SQP and Riemannian gradient methods have nearly identical behavior near the constraint manifold, which could be of broader interest for understanding constrained optimization.

This page was built for publication: Analysis of Sequential Quadratic Programming through the Lens of Riemannian Optimization

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6301951)