federated
federated copied to clipboard
Asynchronous Federated Learning
Does the TensorFlow Federated include asynchronous federated learning? If not, can this framework be used to implement an asynchronous federated learning system?
There are not currently any asynchronous execution contexts available through TFF. That being said, if your interest is purely in simulations, then you could certainly simulate asynchronous client computations. For example, you could write an algorithm that stores the last t
models held by the server, and where clients download one of them at random.
Closing due to lack of activity.