Gradient Methods with Online Scaling Part II. Practical Aspects
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911194751696896 |
|---|---|
| author | Chu, Ya-Chi Gao, Wenzhi Ye, Yinyu Udell, Madeleine |
| author_facet | Chu, Ya-Chi Gao, Wenzhi Ye, Yinyu Udell, Madeleine |
| contents | Part I of this work [Gao25] establishes online scaled gradient methods (OSGM), a framework that utilizes online convex optimization to adapt stepsizes in gradient methods. This paper focuses on the practical aspects of OSGM. We leverage the OSGM framework to design new adaptive first-order methods and provide insights into their empirical behavior. The resulting method, OSGM-Best, matches the performance of quasi-Newton variants while requiring less memory and cheaper iterations. We also extend OSGM to nonconvex optimization and outline directions that connect OSGM to existing branches of optimization theory and practice. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_11007 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Gradient Methods with Online Scaling Part II. Practical Aspects Chu, Ya-Chi Gao, Wenzhi Ye, Yinyu Udell, Madeleine Optimization and Control Machine Learning Part I of this work [Gao25] establishes online scaled gradient methods (OSGM), a framework that utilizes online convex optimization to adapt stepsizes in gradient methods. This paper focuses on the practical aspects of OSGM. We leverage the OSGM framework to design new adaptive first-order methods and provide insights into their empirical behavior. The resulting method, OSGM-Best, matches the performance of quasi-Newton variants while requiring less memory and cheaper iterations. We also extend OSGM to nonconvex optimization and outline directions that connect OSGM to existing branches of optimization theory and practice. |
| title | Gradient Methods with Online Scaling Part II. Practical Aspects |
| topic | Optimization and Control Machine Learning |
| url | https://arxiv.org/abs/2509.11007 |