Abstract: Secure aggregation is an information-theoretic mechanism for gradient aggregation in federated learning, to aggregate the local user gradients without revealing them in the clear. In this ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results