https://github.com/harry24k/pgd-pytorch

A pytorch implementation of "Towards Deep Learning Models Resistant to Adversarial Attacks"
https://github.com/harry24k/pgd-pytorch

adversarial-attacks deep-learning pytorch

Last synced: about 2 months ago
JSON representation

A pytorch implementation of "Towards Deep Learning Models Resistant to Adversarial Attacks"

• Host: GitHub
• URL: https://github.com/harry24k/pgd-pytorch
• Owner: Harry24k
• License: mit
• Created: 2019-04-17T04:41:07.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2019-09-04T14:46:07.000Z (over 5 years ago)
• Last Synced: 2023-03-04T02:24:36.724Z (almost 2 years ago)
• Topics: adversarial-attacks, deep-learning, pytorch
• Language: Jupyter Notebook
• Size: 621 KB
• Stars: 96
• Watchers: 2
• Forks: 26
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# PGD-pytorch

**A pytorch implementation of "[Towards Deep Learning Models Resistant to Adversarial Attacks](https://arxiv.org/abs/1706.06083)"**

## Summary

This code is a pytorch implementation of **PGD attack**   

In this code, I used above methods to fool [Inception v3](https://arxiv.org/abs/1512.00567).   

'[Giant Panda](http://www.image-net.org/)' used for an example.   

You can add other pictures with a folder with the label name in the 'data/imagenet'.    

## Requirements

* python==3.6   

* numpy==1.14.2   

* pytorch==1.0.1   

## Important results not in the code

- Capacity(size of network) plays an important role in adversarial training. (p.9-10)

	- For only natural examples training, it increases the robustness against one-step perturbations.

	- For PGD adversarial training, small capacity networks fails.

	- As capacity increases, the model can fit the adversairal examples increasingly well.

	- More capacity and strong adversaries decrease transferability. (Section B)

- FGSM adversaries don't increase robustness for large epsilon(=8). (p.9-10)

	- The network overfit to FGSM adversarial examples.

- Adversarial training with PGD shows good enough defense results.(p.12-13)

## Notice

- This Repository won't be updated.

- Please check [the package of adversarial attacks in pytorch](https://github.com/Harry24k/adversairal-attacks-pytorch)