Towards Understanding the Dynamics of the First-Order Adversaries

ICML, pp. 2484-2493, 2020.

Cited by: 0|Views25
EI
Weibo:
We present our main results on the convergence of projected gradient descent

Abstract:

An acknowledged weakness of neural networks is their vulnerability to adversarial perturbations to the inputs. To improve the robustness of these models, one of the most popular defense mechanisms is to alternatively maximize the loss over the constrained perturbations (or called adversaries) on the inputs using projected gradient ascen...More

Code:

Data:

0
Your rating :
0

 

Tags
Comments