The Effects of Image Distribution and Task on Adversarial Robustness
In this paper, we propose an adaptation to the area under the curve (AUC) metric to measure the adversarial robustness of a model over a particular ε-interval [ε_0, ε_1] (interval of adversarial perturbation strengths) that facilitates unbiased comparisons across models when they have different init...
Main Authors: | Kunhardt, Owen, Deza, Arturo, Poggio, Tomaso |
---|---|
Format: | Technical Report |
Published: |
Center for Brains, Minds and Machines (CBMM)
2021
|
Online Access: | https://hdl.handle.net/1721.1/129813 |
Similar Items
-
Hierarchically Local Tasks and Deep Convolutional Networks
by: Deza, Arturo, et al.
Published: (2020) -
Biologically Inspired Mechanisms for Adversarial Robustness
by: Vuyyuru Reddy, Manish, et al.
Published: (2020) -
PCA as a defense against some adversaries
by: Aparne, Gupta, et al.
Published: (2022) -
Between-Class Adversarial Training for Improving Adversarial Robustness of Image Classification
by: Desheng Wang, et al.
Published: (2023-03-01) -
Foveation-based Mechanisms Alleviate Adversarial Examples
by: Lou, Yan, et al.
Published: (2016)