Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2306.05896 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866913312574275584 |
|---|---|
| author | Brouste, Alexandre Esstafa, Youssef |
| author_facet | Brouste, Alexandre Esstafa, Youssef |
| contents | A generic, fast and asymptotically efficient method for parametric estimation is described. It is based on the projected stochastic gradient descent on the log-likelihood function corrected by a single step of the Fisher scoring algorithm. We show theoretically and by simulations that it is an interesting alternative to the usual stochastic gradient descent with averaging or the adaptative stochastic gradient descent. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2306_05896 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | One-step corrected projected stochastic gradient descent for statistical estimation Brouste, Alexandre Esstafa, Youssef Statistics Theory Machine Learning A generic, fast and asymptotically efficient method for parametric estimation is described. It is based on the projected stochastic gradient descent on the log-likelihood function corrected by a single step of the Fisher scoring algorithm. We show theoretically and by simulations that it is an interesting alternative to the usual stochastic gradient descent with averaging or the adaptative stochastic gradient descent. |
| title | One-step corrected projected stochastic gradient descent for statistical estimation |
| topic | Statistics Theory Machine Learning |
| url | https://arxiv.org/abs/2306.05896 |