Skip to content

asu-crypto/ppgan

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

PPGAN

Official Implementation of the paper: "Achieving Data Reconstruction Hardness and Efficient Computation in Multiparty Minimax Training" link

CelebA dataset

Download the cropped 64x64 CelebA dataset for training at link

PPGAN training

How to install:

conda create -n ppgan python=3.10
conda activate ppgan
cd crypten
pip install .
cp default.yaml ~/anaconda3/envs/ppgan/lib/python3.10/site-packages/configs/default.yaml

How to run experiments:

CelebA:

cd ppgan_training
./exp_{EXPERIMENT NUMBER}

Where EXPERIMENT NUMBER indicates which configurations we want to run, experiment 1 indicates plaintext GAN training, experiment 2 indicates private D GAN training, experiment 3,4,5 indicate 3,2,1-layer secure D respectively.

MNIST:

cd ppgan_training
./mnist_{EXPERIMENT NUMBER}

Where EXPERIMENT NUMBER indicates which configurations we want to run, 1 indicates

PPGAN Analysis

How to run the analysis:

cd ppgan_analysis
python main.py -ds celeba -bsr 1 -bsf 1 -ns NUM_SECURE > log.txt

where NUM_SECURE indicates number of secure layer we intend to test, ranging from 1 to 3. with NUM_SECURE=0, we have Federated Learning protocol where no layer in the network are secure.

About

Official Implementation of the paper: "Achieving Data Reconstruction Hardness and Efficient Computation in Multiparty Minimax Training"

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors