Adversarial-Attacks-against-Neural-Networks

In this repository I give a brief introduction to the topic. I also generate some adversarial examples using Fast gradient sign method. Then I use those images to attack a trained network. Lastly I use adversarial examples generated for one neural network to attack another neural network. This is used to show that adversarial attacks are transferable from one network to another