Filters








1 Hit in 1.9 sec

On the Security of Randomized Defenses Against Adversarial Samples [article]

Kumar Sharad, Giorgia Azzurra Marson, Hien Thi Thu Truong, Ghassan Karame
2020 arXiv   pre-print
In this paper, we study the effectiveness of randomized defenses against adversarial samples.  ...  Among these, a promising approach is to use randomness in order to make the classification process unpredictable and presumably harder for the adversary to control.  ...  We note that Randomized Squeezing, while using squeezing routines, does not aim at detecting adversarial samples, rather at making it harder for the adversary to generate successful perturbations.  ... 
arXiv:1812.04293v4 fatcat:q6xrfhzxpbgqtldvie2qlbvcru