search

wenzhu23333 / Differential-Privacy-Based-Federated-Learning

Everything you want about DP-Based Federated Learning, including Papers and Code. (Mechanism: Laplace or Gaussian, Dataset: femnist, shakespeare, mnist, cifar-10 and fashion-mnist. )
GNU General Public License v3.0
348 stars 56 forks source link

Hyperparameters for training cnn for cifar 10 #13

Closed Toby-Kang closed 1 year ago

Toby-Kang commented 1 year ago

Hi Wenzhu! I wonder if you can help to share the hyperparameters you used when training the CNN model for cifar10 dataset. For me the best test accuracy that I can achieve when there is no dp is around 34%. Is it the similar story on your side as well? Many thanks!

wenzhu23333 commented 1 year ago

The following are the hyperparameters and results I once trained: For dp-sgd： dp_clip=1e-5, frac=1, sampling rate=0.1, client num=50, lr=0.2, lr_decay=0.999, 2000 rounds no_dp: 69% epsilon=350: 52% epsilon=250: 51%