Inexact Adaptive Cubic Regularization Algorithms on Riemannian Manifolds and Application
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_ | 1866916235407523840 |
|---|---|
| author | Li, Z. Y. Wang, X. M. |
| author_facet | Li, Z. Y. Wang, X. M. |
| contents | The adaptive cubic regularization algorithm employing the inexact gradient and Hessian is proposed on general Riemannian manifolds, together with the iteration complexity to get an approximate second-order optimality under certain assumptions on accuracies about the inexact gradient and Hessian. The algorithm extends the inexact adaptive cubic regularization algorithm under true gradient in [Math. Program., 184(1-2): 35-70, 2020] to more general cases even in Euclidean settings. As an application, the algorithm is applied to solve the joint diagonalization problem on the Stiefel manifold. Numerical experiments illustrate that the algorithm performs better than the inexact trust-region algorithm in [Advances of the neural information processing systems, 31, 2018]. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_02588 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Inexact Adaptive Cubic Regularization Algorithms on Riemannian Manifolds and Application Li, Z. Y. Wang, X. M. Optimization and Control Numerical Analysis 53C20(Primary), 53C22(Secondary) The adaptive cubic regularization algorithm employing the inexact gradient and Hessian is proposed on general Riemannian manifolds, together with the iteration complexity to get an approximate second-order optimality under certain assumptions on accuracies about the inexact gradient and Hessian. The algorithm extends the inexact adaptive cubic regularization algorithm under true gradient in [Math. Program., 184(1-2): 35-70, 2020] to more general cases even in Euclidean settings. As an application, the algorithm is applied to solve the joint diagonalization problem on the Stiefel manifold. Numerical experiments illustrate that the algorithm performs better than the inexact trust-region algorithm in [Advances of the neural information processing systems, 31, 2018]. |
| title | Inexact Adaptive Cubic Regularization Algorithms on Riemannian Manifolds and Application |
| topic | Optimization and Control Numerical Analysis 53C20(Primary), 53C22(Secondary) |
| url | https://arxiv.org/abs/2405.02588 |