METHOD AND APPARATUS FOR TRAINING ADVERSARIAL NETWORK, METHOD AND APPARATUS FOR SUPPLEMENTING MEDICAL DATA, AND DEVICE AND MEDIUM

摘要

The present application relates to the technical field of artificial intelligence and is applied to the field of smart medical treatment. Disclosed are a method and apparatus for training an adversarial network, a method and apparatus for supplementing medical data, and a device and a medium. The method for training an adversarial network comprises: acquiring an initial adversarial network, wherein the initial adversarial network comprises a generator model that includes an initial parameter, and a trained sensing network model; inputting a preset random noise into the initial adversarial network, and generating generated data corresponding to the random noise by means of the generator model; determining, according to the generated data, a total loss value of the generator model by means of the sensing network model; and when the total loss value does not reach a preset convergence condition, updating and iterating the initial parameter of the generator model until the total loss value reaches the preset convergence condition, and recording the initial adversarial network after convergence as an adversarial network. According to the present application, by improving a GAN, the function of an adversarial network obtained by means of training is expanded, and the data generation accuracy of the adversarial network is improved.