Solving nonlinear systems of equations via spectral residual methods:\n stepsize selection and applications

2020/05/12 by Enrico Meli, Benedetta Morini, Meli, Enrico +5
Computer Science · Decision Sciences · Mathematics · #Advanced Optimization Algorithms Research #FOS: Mathematics #Metaheuristic Optimization Algorithms Research #Numerical Analysis (math.NA) #Optimization and Control (math.OC) #Scheduling and Timetabling Solutions

paper · pdf · doi:10.48550/arxiv.2005.05851

openalex publication_date 2020/05/12 · openalex created_date 2022/07/26 · openalex updated_date 2026/07/28

Abstract

Spectral residual methods are derivative-free and low-cost per iteration\nprocedures for solving nonlinear systems of equations. They are generally\ncoupled with a nonmonotone linesearch strategy and compare well with\nNewton-based methods for large nonlinear systems and sequences of nonlinear\nsystems. The residual vector is used as the search direction and choosing the\nsteplength has a crucial impact on the performance. In this work we address\nboth theoretically and experimentally the steplength selection and provide\nresults on a real application such as a rolling contact problem.\n

Related

• Accelerated derivative-free spectral residual method for nonlinear systems of equations
• Spectral Residual Method for Nonlinear Equations on Riemannian Manifolds
• Minimal residual multistep methods for large stiff non-autonomous linear problems
• Various approaches to solving nonlinear equations
• Residual subspace evolution strategies for nonlinear inverse problems