RANCER: non-axis aligned anisotropic certification with randomized smoothing
As modern networks have been proven to be unprotected from adversarial attacks and are applied in safety-critical applications, defense against them is very crucial. Many works were dedicated to this topic, but randomized smoothing has been recently proven to be an effective approach for the certifi...
Main Authors: | Rumezhak, T, Eiras, FG, Torr, PHS, Bibi, A |
---|---|
Format: | Conference item |
Language: | English |
Published: |
IEEE
2023
|
Similar Items
-
Certifying ensembles: a general certification theory with s-lipschitzness
by: Petrov, A, et al.
Published: (2023) -
Towards certification of uncertainty calibration under adversarial attacks
by: Emde, C, et al.
Published: (2024) -
Data dependent randomized smoothing
by: Alfarra, M, et al.
Published: (2022) -
Tight certificates of adversarial robustness for randomly smoothed classifiers
by: Lee, Guang-He, et al.
Published: (2021) -
DeformRS: Certifying input deformations with randomized smoothing
by: Alfarra, M, et al.
Published: (2022)