Skip to content

ApurbaSengupta/Fast-Gradient-Sign-Attack

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 

Repository files navigation

Fast-Gradient-Sign-Attack

Attacking a Convolutional Neural Network (CNN) having ReLU activations with adversarial images generated using Fast Gradient Sign Attack (FGSM) techniques (both un-targeted and targeted), thus exposing threats to a highly confident CNN.

Utilized PyTorch framework for development. Used Google Colaboratory to use GPU while training the model.