Abstract
This paper presents a new iterative solver for least-squares problems, which is developed based on the Levenberg-Marquardt and trust region methods. The proposed iterative solver substantially reduces computing time compared with the conventional Levenberg-Marquardt scheme, and can be efficiently applied to large-scale problems. Three numerical examples are used to demonstrate the effectiveness of the proposed solver.
| Original language | English |
|---|---|
| Pages (from-to) | 327-335 |
| Number of pages | 9 |
| Journal | Communications in Numerical Methods in Engineering |
| Volume | 21 |
| Issue number | 6 |
| DOIs | |
| Publication status | Published - Jun 2005 |
| Externally published | Yes |
Keywords
- Iterative solver
- Least-squares problem
- Levenberg-Marquardt
- Trust region