DAM-SE: A Blockchain-Based Optimized Solution for the Counterattacks in the Internet of Federated Learning Systems

Algorithm 1

Model update of worker in round .

	Input: model updates submitted by all clusters in round e − 1, or
	Output: trained model parameters , Voting results
	/∗1. Use local data to select the best a model to update (except the first round)∗/
(1)	ifthen
(2)	fordo
(3)
(4)	end
(5)	⟵ chose a models whose is the smallest.
(6)	end
	/∗2. Aggregation with a models value ∗/
(7)	ifthen
(8)
(9)	else
(10)
(11)	end
	/∗ Update the model with local data ∗/
(12)
(13)
(14)	for each local epochs E do
(15)	for each batch b in do
(16)
(17)	end
(18)	end
(19)	return,