Some convergence properties of the steepest descent algorithm revealed by renormalisation.

MaRDI QIDQ2768064zbMATH OpenFDO

Authors Luc Pronzato, Henry P. Wynn, A. Zhigljavsky

Publication date 2001

zbMATH Keywords

asymptotic behaviourattractorrelaxation

Mathematics Subject Classification ID

Quadratic programming (90C20) Methods of reduced gradient type (90C52)

Recommendations

• Renormalised steepest descent in Hilbert space converges to a two-point attractor.
• On the steepest descent algorithm for quadratic functions
• On the worst case performance of the steepest descent algorithm for quadratic functions
• On the convergence of descent algorithms
• Steepest descent method with random step lengths

Cited in

(5)

• Renormalised steepest descent in Hilbert space converges to a two-point attractor.
• The chaotic nature of faster gradient descent methods
• On initial point selection of the steepest descent algorithm for general quadratic functions
• On the behavior of the gradient norm in the steepest descent method
• Studying the rate of convergence of the steepest descent optimisation algorithm with relaxation

This page was built for publication: Some convergence properties of the steepest descent algorithm revealed by renormalisation.

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2768064)