Genetic CFL: Hyperparameter Optimization in Clustered Federated Learning

Algorithm 2

Genetic optimization based FL on clustered data.

	rounds: Number of loops for training the federated model
(1)	function Mutate ()
(2)	factor random ()
(3)
(4)	return
(5)	function Crossover ()
(6)	Initialize temporary array to store
(7)
(8)	for to do
(9)
(10)
(11)	return
(12)	function Evolve ()
(13)	,
(14)
(15)	return Crossover ()
(16)	procedure Train
(17)
(18)	to len
(19)	Initialize with shape (len, size (clusters )
(20)
(21)	for to rounds do
(22)	for to do
(23)	ind = [clusters.index (cluster[i])]
(24)
(25)	Empty arrays losses,
(26)	for to n do
(27)	, losses[k],
(28)