A structured L-BFGS method with diagonal scaling and its application to image registration
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917771145641984 |
|---|---|
| author | Mannel, Florian Aggrawal, Hari Om |
| author_facet | Mannel, Florian Aggrawal, Hari Om |
| contents | We devise an L-BFGS method for optimization problems in which the objective is the sum of two functions, where the Hessian of the first function is computationally unavailable while the Hessian of the second function has a computationally available approximation that allows for cheap matrix-vector products. This is a prototypical setting for many inverse problems. The proposed L-BFGS method exploits the structure of the objective to construct a more accurate Hessian approximation than in standard L-BFGS. In contrast to existing works on structured L-BFGS, we choose the first part of the seed matrix, which approximates the Hessian of the first function, as a diagonal matrix rather than a multiple of the identity. We derive two suitable formulas for the coefficients of the diagonal matrix and show that this boosts performance on real-life image registration problems, which are highly non-convex inverse problems. The new method converges globally and linearly on non-convex problems under mild assumptions in a general Hilbert space setting, making it applicable to a broad class of inverse problems. An implementation of the method is freely available. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_19834 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A structured L-BFGS method with diagonal scaling and its application to image registration Mannel, Florian Aggrawal, Hari Om Optimization and Control 65J22 65K05 65K10 90C06 90C26 90C30 90C48 90C53 90C90 We devise an L-BFGS method for optimization problems in which the objective is the sum of two functions, where the Hessian of the first function is computationally unavailable while the Hessian of the second function has a computationally available approximation that allows for cheap matrix-vector products. This is a prototypical setting for many inverse problems. The proposed L-BFGS method exploits the structure of the objective to construct a more accurate Hessian approximation than in standard L-BFGS. In contrast to existing works on structured L-BFGS, we choose the first part of the seed matrix, which approximates the Hessian of the first function, as a diagonal matrix rather than a multiple of the identity. We derive two suitable formulas for the coefficients of the diagonal matrix and show that this boosts performance on real-life image registration problems, which are highly non-convex inverse problems. The new method converges globally and linearly on non-convex problems under mild assumptions in a general Hilbert space setting, making it applicable to a broad class of inverse problems. An implementation of the method is freely available. |
| title | A structured L-BFGS method with diagonal scaling and its application to image registration |
| topic | Optimization and Control 65J22 65K05 65K10 90C06 90C26 90C30 90C48 90C53 90C90 |
| url | https://arxiv.org/abs/2405.19834 |