Federated learning excitation method and system based on differential evolution

The embodiment of the invention provides a federated learning excitation method and system based on differential evolution, and the method comprises the steps that all participants transfer along withtime t (assuming that a month is a period) in a federated learning process through the excellent global optimization capability and local detection capability of DE; the difference (expected loss) between the earnings obtained by each participant from the federation and the earnings which should be obtained by each participant minimizes the expected loss and waiting time between the participants,and automatically balances the actual earnings of each participant in federation learning and the obtained return expected difference; the participant is effectively promoted to provide reliable data, so that federated learning can be carried out stably for a long time. The method effectively achieves the dynamic adjustment of the federated learning total earnings and the earnings of each participant, maximizes the sustai