Existence and Minimax Theorems for Adversarial Surrogate Risks in Binary Classification
Fuente:
arXiv
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2022
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911196190343168 |
|---|---|
| author | Frank, Natalie S. Niles-Weed, Jonathan |
| author_facet | Frank, Natalie S. Niles-Weed, Jonathan |
| contents | Adversarial training is one of the most popular methods for training methods robust to adversarial attacks, however, it is not well-understood from a theoretical perspective. We prove and existence, regularity, and minimax theorems for adversarial surrogate risks. Our results explain some empirical observations on adversarial robustness from prior work and suggest new directions in algorithm development. Furthermore, our results extend previously known existence and minimax theorems for the adversarial classification risk to surrogate risks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2206_09098 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Existence and Minimax Theorems for Adversarial Surrogate Risks in Binary Classification Frank, Natalie S. Niles-Weed, Jonathan Machine Learning Statistics Theory Adversarial training is one of the most popular methods for training methods robust to adversarial attacks, however, it is not well-understood from a theoretical perspective. We prove and existence, regularity, and minimax theorems for adversarial surrogate risks. Our results explain some empirical observations on adversarial robustness from prior work and suggest new directions in algorithm development. Furthermore, our results extend previously known existence and minimax theorems for the adversarial classification risk to surrogate risks. |
| title | Existence and Minimax Theorems for Adversarial Surrogate Risks in Binary Classification |
| topic | Machine Learning Statistics Theory |
| url | https://arxiv.org/abs/2206.09098 |