Enhancing Sharpness-Aware Minimization by Learning Perturbation Radius
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913468741844992 |
|---|---|
| author | Wang, Xuehao Jiang, Weisen Fu, Shuai Zhang, Yu |
| author_facet | Wang, Xuehao Jiang, Weisen Fu, Shuai Zhang, Yu |
| contents | Sharpness-aware minimization (SAM) is to improve model generalization by searching for flat minima in the loss landscape. The SAM update consists of one step for computing the perturbation and the other for computing the update gradient. Within the two steps, the choice of the perturbation radius is crucial to the performance of SAM, but finding an appropriate perturbation radius is challenging. In this paper, we propose a bilevel optimization framework called LEarning the perTurbation radiuS (LETS) to learn the perturbation radius for sharpness-aware minimization algorithms. Specifically, in the proposed LETS method, the upper-level problem aims at seeking a good perturbation radius by minimizing the squared generalization gap between the training and validation losses, while the lower-level problem is the SAM optimization problem. Moreover, the LETS method can be combined with any variant of SAM. Experimental results on various architectures and benchmark datasets in computer vision and natural language processing demonstrate the effectiveness of the proposed LETS method in improving the performance of SAM. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_08222 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Enhancing Sharpness-Aware Minimization by Learning Perturbation Radius Wang, Xuehao Jiang, Weisen Fu, Shuai Zhang, Yu Machine Learning Sharpness-aware minimization (SAM) is to improve model generalization by searching for flat minima in the loss landscape. The SAM update consists of one step for computing the perturbation and the other for computing the update gradient. Within the two steps, the choice of the perturbation radius is crucial to the performance of SAM, but finding an appropriate perturbation radius is challenging. In this paper, we propose a bilevel optimization framework called LEarning the perTurbation radiuS (LETS) to learn the perturbation radius for sharpness-aware minimization algorithms. Specifically, in the proposed LETS method, the upper-level problem aims at seeking a good perturbation radius by minimizing the squared generalization gap between the training and validation losses, while the lower-level problem is the SAM optimization problem. Moreover, the LETS method can be combined with any variant of SAM. Experimental results on various architectures and benchmark datasets in computer vision and natural language processing demonstrate the effectiveness of the proposed LETS method in improving the performance of SAM. |
| title | Enhancing Sharpness-Aware Minimization by Learning Perturbation Radius |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2408.08222 |