An Adaptive Cubic Regularization quasi-Newton Method on Riemannian Manifolds
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_ | 1866910336440860672 |
|---|---|
| author | Louzeiro, Mauricio S. Silva, Gilson N. Yuan, Jinyun Zhang, Daoping |
| author_facet | Louzeiro, Mauricio S. Silva, Gilson N. Yuan, Jinyun Zhang, Daoping |
| contents | A quasi-Newton method with cubic regularization is designed for solving Riemannian unconstrained nonconvex optimization problems. The proposed algorithm is fully adaptive with at most ${\cal O} (ε_g^{-3/2})$ iterations to achieve a gradient smaller than $ε_g$ for given $ε_g$, and at most $\mathcal O(\max\{ ε_g^{-\frac{3}{2}}, ε_H^{-3} \})$ iterations to reach a second-order stationary point respectively. Notably, the proposed algorithm remains applicable even in cases of the gradient and Hessian of the objective function unknown. Numerical experiments are performed with gradient and Hessian being approximated by forward finite-differences to illustrate the theoretical results and numerical comparison. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_12464 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | An Adaptive Cubic Regularization quasi-Newton Method on Riemannian Manifolds Louzeiro, Mauricio S. Silva, Gilson N. Yuan, Jinyun Zhang, Daoping Optimization and Control A quasi-Newton method with cubic regularization is designed for solving Riemannian unconstrained nonconvex optimization problems. The proposed algorithm is fully adaptive with at most ${\cal O} (ε_g^{-3/2})$ iterations to achieve a gradient smaller than $ε_g$ for given $ε_g$, and at most $\mathcal O(\max\{ ε_g^{-\frac{3}{2}}, ε_H^{-3} \})$ iterations to reach a second-order stationary point respectively. Notably, the proposed algorithm remains applicable even in cases of the gradient and Hessian of the objective function unknown. Numerical experiments are performed with gradient and Hessian being approximated by forward finite-differences to illustrate the theoretical results and numerical comparison. |
| title | An Adaptive Cubic Regularization quasi-Newton Method on Riemannian Manifolds |
| topic | Optimization and Control |
| url | https://arxiv.org/abs/2402.12464 |