Federated Learning Framework Based on Data Value Evaluation in Industrial IoT

Table 1

The accuracy of the global model on the validation set.

Set		Methods
		FedAvg	DVRL	FLDVE
MNIST	Clean	0.9960	—	—
	Set-1	0.990	0.984	0.992
	Set-2	0.982	0.978	0.990
	Set-3	0.974	0.970	0.986
CIFAR-10	Clean	0.7929	—	—
	Set-1	0.7589	0.7329	0.7747
	Set-2	0.6829	0.6722	0.7010
	Set-3	0.6778	0.6429	0.6913

The numbers in the table refer to the accuracy of the model in the test set. The bold values mark the maximum accuracy in a certain set.