Skip to content

Repository consists of pre-trained CNN model in pytorch, hitting 89% on Fashion MNIST dataset. Adversarial attack was implemented on a given model. Results are below.

Notifications You must be signed in to change notification settings

makhanov/CNN_pytorch_adversarial_attack_Fashion_MNIST

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

CNN_pytorch evasion attack on Fashion MNIST

This repository consists of my implementation of Adversarial CNN model in pytorch, hitting 89.22% on Fashion-MNIST test dataset. Trained model also was provided in order to run the attack. Based on our trained model we implemented evasion attack (FGSM). Results are:

Epsilon: 0 Test Accuracy = 8922 / 10000 = 0.8922 /n

Epsilon: 0.05 Test Accuracy = 3595 / 10000 = 0.3595

Epsilon: 0.1 Test Accuracy = 1663 / 10000 = 0.1663

Epsilon: 0.15 Test Accuracy = 784 / 10000 = 0.0784

Epsilon: 0.2 Test Accuracy = 361 / 10000 = 0.0361

Epsilon: 0.25 Test Accuracy = 230 / 10000 = 0.023

Epsilon: 0.3 Test Accuracy = 181 / 10000 = 0.0181

In folder 'saved' you can see the result figure 'accuracy vs epsilon':

where the higher epsilon is, the lower the accuracy of our model. Next I would like to read papers where CNN model accuracy is higher than 90%, retrain my model and rerun again the attack.

About

Repository consists of pre-trained CNN model in pytorch, hitting 89% on Fashion MNIST dataset. Adversarial attack was implemented on a given model. Results are below.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages