WCL: Client Selection in Federated Learning with a Combination of Model Weight Divergence and Client Training Loss for Internet Traffic Classification

<div>An overview of the proposed algorithm WCL. It can be seen from the figure that WCL has two more steps compared to FL in which all clients participate: (a) calculate weight divergence (bias) and obtain client loss and (b) client selection.</div>

Wireless Communications and Mobile Computing

fig2

Figure 2

Figure 2: WCL: Client Selection in Federated Learning with a Combination of Model Weight Divergence and Client Training Loss for Internet Traffic Classification 

Figure 2 | WCL: Client Selection in Federated Learning with a Combination of Model Weight Divergence and Client Training Loss for Internet Traffic Classification 