An Adaptive Communication-Efficient Federated Learning to Resist Gradient-Based Reconstruction Attacks

Algorithm 1

An adaptive communication-efficient federated learning (AFC-FL).

	Input: The number of clients ;
	Input: Training epochs ;
	Input: Updating epochs ;
	Input: Learning rate ;
	Input: Local mini-batch size ;
	Input: Local optimization function Adam;
	Output: A global model ;
(1)	Procedure ServerExecute:
(2)	initialize ;
(3)	for each Iteration do
(4)	;
(5)	end for
(6)	end procedure
(7)	Procedure ClientUpdate (m):
(8)	for each Iteration do
(9)	ifthen
(10)
(11)
(12)	end if
(13)	for each Client do
(14)
(15)
(16)	end for
(17)	end for
(18)	end procedure