Generative Adversarial Networks – a clever setup where two neural networks compete against each other. One generates fakes, the other spots fakes, and through this eternal battle, they both get better. Like siblings who constantly try to outsmart each other.
Synonyms:
Generative Adversarial Networks