Skip to content

Latest commit

 

History

History
39 lines (26 loc) · 1.24 KB

README.md

File metadata and controls

39 lines (26 loc) · 1.24 KB
Raw

README

This is for releasing the source code of the paper "RamBoAttack: A Robust Query Efficient Deep Neural Network Decision Exploit"

Archived Version: RamBoAttack

The project is published as part of the following paper and if you re-use our work, please cite the following paper:

@inproceedings{vo2022,
title={RamBoAttack: A Robust Query Efficient Deep Neural Network Decision Exploit},
author={Viet Quoc Vo and Ehsan Abbasnejad and Damith C. Ranasinghe},
year = {2022},
journal = {Network and Distributed Systems Security (NDSS) Symposium},
}

The source code is written mostly on Python 3 and Pytorch, so please help to download and install Python3 and Pytorch beforehand.

Requirements

To install the requirements for this repo, run the following command:

git clone https://github.com/RamBoAttack/RamBoAttack.github.io.git
cd RamBoAttack
pip3 install -r requirements.txt

Run the RamBoAttack

  • Run step-by-step with the Jupyter Notebook file 'Tutorial - RamBoAttack - CIFAR10.ipynb' or 'Tutorial - RamBoAttack - ImageNet.ipynb'.

TODO

  • add the testing code.

Notes:

  1. The pretrained model for CIFAR-10 can be downloaded from this repo.