A Generalization Bound for Nearly-Linear Networks
Fuente:
arXiv
Guardado en:
| Autor principal: | |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866913423149760512 |
|---|---|
| author | Golikov, Eugene |
| author_facet | Golikov, Eugene |
| contents | We consider nonlinear networks as perturbations of linear ones. Based on this approach, we present novel generalization bounds that become non-vacuous for networks that are close to being linear. The main advantage over the previous works which propose non-vacuous generalization bounds is that our bounds are a-priori: performing the actual training is not required for evaluating the bounds. To the best of our knowledge, they are the first non-vacuous generalization bounds for neural nets possessing this property. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_06765 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A Generalization Bound for Nearly-Linear Networks Golikov, Eugene Machine Learning Artificial Intelligence We consider nonlinear networks as perturbations of linear ones. Based on this approach, we present novel generalization bounds that become non-vacuous for networks that are close to being linear. The main advantage over the previous works which propose non-vacuous generalization bounds is that our bounds are a-priori: performing the actual training is not required for evaluating the bounds. To the best of our knowledge, they are the first non-vacuous generalization bounds for neural nets possessing this property. |
| title | A Generalization Bound for Nearly-Linear Networks |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2407.06765 |