REMARKS ON THE GRADIENT-PROJECTION ALGORITHM

Authors

  • M. Su PENN STATE UNIVERSITY AT ERIE , THE BEHREND COLLEGE, 4205 COLLEGE DRIVE, ERIE , PA 16563 - 0203, USA
  • H. Xu DEPARTMENT OF APPLIED MATHEMATICS , NATIONAL SUNYATSEN UNIVERSITY, KAOHSIUNG 80424, TAIWAN

Abstract

The gradient-projection algorithm (GPA) is a powerful method for solving constrained minimization problems infinite (and even infinite) dimensional Hilbert spaces. We consider GPA with variable stepsizes and show that if GPA generates a bounded sequence, then under certain assumptions, every accumulation point of the sequence is a solution of the minimization problem. We also look into the issue where the sequence of step sizes is allowed to be the limiting case (e.g., approaching to zero).

Downloads

Published

2011-12-18

How to Cite

Su, M., & Xu, H. (2011). REMARKS ON THE GRADIENT-PROJECTION ALGORITHM. Journal of Nonlinear Analysis and Optimization: Theory & Applications, 1(1), 35-43. Retrieved from http://www.math.sci.nu.ac.th/ojs302/index.php/jnao/article/view/59

Issue

Section

Research Article