Open AccessOn damping parameters of Levenberg-Marquardt algorithm for nonlinear least square problems
Author(s) -
Abubakar Umar,
Ibrahim Mohammed Sulaiman,
Mustafa Mamat,
Mohammed Yusuf Waziri,
Nurnadiah Zamri
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1734/1/012018
Subject(s) - levenberg–marquardt algorithm , nonlinear system , scaling , algorithm , square (algebra) , non linear least squares , line search , mathematical optimization , mathematics , computer science , estimation theory , artificial intelligence , artificial neural network , physics , geometry , quantum mechanics , computer security , radius
The Levenberg-Marquardt (LM) algorithm is a widely used method for solving problems related to nonlinear least squares. The method depends on a nonlinear parameter μ known as self-scaling parameter that affects the performance of the algorithm. In this paper we examine the effect of various choice of parameters and of relaxing the line search. Numerical results obtained are used to compare the performance using standard test problems which show that the proposed alternatives are promising.