Skip to content

Privacy-Preserving Deep Learning via Additively Homomorphic Encryption

Notifications You must be signed in to change notification settings

lucifer2859/Paillier-LWE-based-PHE

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

75 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Paillier-LWE-based-PHE

Privacy-Preserving Deep Learning via Additively Homomorphic Encryption

CUDA版本:9.0与9.1皆可

PyTorch: 1.1.0

python-paillier-master

python setup.py test

python test.py

LWE-based PHE

mkdir key // 创建用于存储密钥的文件夹

python cpu_test.py | python cuda_test.py

数据集分割

python split_data.py

LeNet训练

mkdir models // 创建用于存储模型的文件夹

LeNet.py、LeNet_subset.py与jointly_learning_demo.py可以独立运行(注意数据集路径)

jointly_learning_with_encryption_demo.py需要注意import LeNet与cuda_test时的路径(可以将其与LeNet.py放入LWE-based PHE根目录中)

代码版本区别

jointly_learning: 上传梯度,下载聚合后的梯度

v1: IID,平衡,数据集没有实际分割,在训练时模拟分割;

v2: IID,平衡,数据集有实际分割;

v3: IID,将用户进行了类封装,每个epoch迭代最小数据集的iteration数,逐用户测试;

v4: 在v3基础上添加了对Non-IID设置以及完整数据集测试的支持;

federated_learning: 上传模型,下载聚合后的模型

v1: FedAvg;

v2: EASGD(Elastic Averaging SGD);

v3: FedProx;

About

Privacy-Preserving Deep Learning via Additively Homomorphic Encryption

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages