Training a federated generative adversarial network

A computer-implemented method, the method comprising:

• training a federated generative adversarial network (GAN) using private data at a participant system having a local feature extractor and a local discriminator, wherein the participant system is in communication with an aggregator system having a generator and a discriminator, the training comprising:
• training a feature extractor to extract a set of features for input to the discriminator at the aggregator system, wherein the features include features extracted from private data that is private to the participant system;
• and
• training a local discriminator to produce discriminator parameter updates to update the discriminator at the aggregator system.

A system comprising:

• a processor;
• and
• a memory in communication with the processor, the memory containing program instructions that, when executed by the processor, are configured as one or more components to cause the processor to perform a method, the one or more components comprising:
• an aggregator system having a generator and a discriminator with an input collector wherein the input collector is in communication with multiple participant systems each having a local feature extractor and a local discriminator;
• the input collector including a feature component for receiving, from the feature extractors at participant systems, a set of features for input to the discriminator, wherein the features include features extracted from private data that is private to a participant system;
• the input collector including a discriminator update component for receiving parameter updates from the local discriminators at participant systems that are trained using local real data and fake data, with the fake data provided by the generator.