Competition on robust deep learning

This perspective paper proposes a new adversarial training method based on large-scale pre-trained models to achieve state-of-the-art adversarial robustness on ImageNet.

Detalles Bibliográficos
Autores principales: Dong, Yinpeng, Liu, Chang, Xiang, Wenzhao, Su, Hang, Zhu, Jun
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Oxford University Press 2023
Materias:
PERSPECTIVE
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10257479/
https://www.ncbi.nlm.nih.gov/pubmed/37304460
http://dx.doi.org/10.1093/nsr/nwad087
Descripción
Sumario:This perspective paper proposes a new adversarial training method based on large-scale pre-trained models to achieve state-of-the-art adversarial robustness on ImageNet.

Ejemplares similares