FEDERATED LEARNING MODEL TRAINING METHOD AND APPARATUS, TERMINAL DEVICE, AND STORAGE MEDIUM

摘要

The present application relates to the technical field of artificial intelligence, and in particular relates to a federated learning model training method and apparatus, a terminal device, and a storage medium. The method comprises: storing initial model parameters of a federated learning model in a blockchain, and respectively instructing participating devices to perform local model training according to the initial model parameters in the blockchain to obtain model update parameters; respectively acquiring the model update parameters from different participating devices, and performing parameter detection on the model update parameters; if the parameter detection of the model update parameters is up to standard, storing the model update parameters in the blockchain; and performing model training on the federated learning model according to the model update parameters in the blockchain until the federated learning model after being subjected to model training converges. In the present application, by means of storing the model update parameters, the parameter detection of which is up to standard, in the blockchain, the storage of false parameters in the blockchain is prevented, and malicious attack of the participating devices on the federated learning model is prevented, thereby improving the accuracy and the security of training of the federated learning model.