Gradient-Based Sampling Federated XGBoost

Hi everyone,

What happens (in practice and theoretically) when enabling gradient_based subsampling for federated learning?

From my understanding, gradient based subsampling is based on Mimimal Variance Sampling (MVS). My questions are:

  1. Is the subsampling based on gradient information from each client or global information?
  2. Can it be used for adaptation to local client data?
  3. Can we prove convergence using MVS in a federated setting?

Hope to start a great discussion!