VasundharaAgarwal

#### Reference Issues/PRs #### What does this implement/fix? Explain your changes. Implements wrappers for Strategy and NumPyClient which can be used to add central DP to the training process. ####...

### What is your question? Hi, I noticed that in the simulation_pytorch example, we pass a function with return type `CifarRayClient` as the `client_fn` parameter in the call to `start_simulation`...

[differential_privacy] Learning rates used for Adaptive Clipping experiments

6

Hi, I am trying to reproduce the experiments in ["Differentially Private Learning with Adaptive Clipping" (2021)](https://arxiv.org/abs/1905.03871), the source code for which is provided under `federated/differential_privacy`. The paper does not report...