Fedrated learning system and method using data digest

A federated learning method using data digest includes: sending a server model to a plurality of client devices by a moderator, generating encoded features according to raw data and performing a training procedure by each client device, wherein the training procedure includes "updating the server model to generate a client model, selecting at least two encoded features and at least two labels to compute a feature weighted sum and a label weighted sum when receiving a digest request, sending a digest including the feature weighted sum and the label weighted sum and update parameters of the client model to the moderator", and performing the following steps by the moderator: "determining an absent client and an available client in the client device, generating a replacement model according to the general model and the absent client, generating an aggregation model according to the available client and the replacement model, and training the aggregation model to update the server model.