Provably effective detection of effective data poisoning attacks
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909462814523392 |
|---|---|
| author | Gallagher, Jonathan Esfandiari, Yasaman MacPhee, Callen Warren, Michael |
| author_facet | Gallagher, Jonathan Esfandiari, Yasaman MacPhee, Callen Warren, Michael |
| contents | This paper establishes a mathematically precise definition of dataset poisoning attack and proves that the very act of effectively poisoning a dataset ensures that the attack can be effectively detected. On top of a mathematical guarantee that dataset poisoning is identifiable by a new statistical test that we call the Conformal Separability Test, we provide experimental evidence that we can adequately detect poisoning attempts in the real world. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_11795 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Provably effective detection of effective data poisoning attacks Gallagher, Jonathan Esfandiari, Yasaman MacPhee, Callen Warren, Michael Cryptography and Security Computer Vision and Pattern Recognition Machine Learning This paper establishes a mathematically precise definition of dataset poisoning attack and proves that the very act of effectively poisoning a dataset ensures that the attack can be effectively detected. On top of a mathematical guarantee that dataset poisoning is identifiable by a new statistical test that we call the Conformal Separability Test, we provide experimental evidence that we can adequately detect poisoning attempts in the real world. |
| title | Provably effective detection of effective data poisoning attacks |
| topic | Cryptography and Security Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2501.11795 |