626_privacy/tensorflow_privacy
2
0
Fork 1
Code Issues Pull requests 1 Projects Releases Packages Wiki Activity Actions

Add citation

Browse source
This commit is contained in:
Nicholas Carlini 2021-12-14 00:59:37 +00:00
parent d2fda3ce1d
commit bf9a58d26b
1 changed files with 16 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

17
research/instahide_attack_2020/README.md
View file

@ -1,6 +1,6 @@
Implementation of our reconstruction attack on InstaHide.
An Attack on InstaHide: Is Private Learning Possible with Instance Encoding?
Is Private Learning Possible with Instance Encoding?
Nicholas Carlini, Samuel Deng, Sanjam Garg, Somesh Jha, Saeed Mahloujifar, Mohammad Mahmoody, Shuang Song, Abhradeep Thakurta, Florian Tramer
https://arxiv.org/abs/2011.05315
@ -49,3 +49,18 @@ To reproduce our results and run the attack, each of the files should be run in
6. Run `step_6_adjust_color.py`. Adjust the color curves to match.
7. Run `step_7_visualize.py`. Show the final resulting images.
## Citation
You can cite this attack at
```
@inproceedings{carlini2021private,
title={Is Private Learning Possible with Instance Encoding?},
author={Carlini, Nicholas and Deng, Samuel and Garg, Sanjam and Jha, Somesh and Mahloujifar, Saeed and Mahmoody, Mohammad and Thakurta, Abhradeep and Tram{\`e}r, Florian},
booktitle={2021 IEEE Symposium on Security and Privacy (SP)},
pages={410--427},
year={2021},
organization={IEEE}
}
```