GAN (Generative Adversarial Network)
/ɡæn/
GAN (Generative Adversarial Network): A framework where two neural networks compete: a generator creates fake data and a discriminator tries to distinguish real from fake. GANs consist of a generator and discriminator trained adversarially. The generator improves at creating realistic data while the discriminator improves at detection.
In italiano: GAN (Generative Adversarial Network)Legge II · Probabilità, softmax e modelli generativiGANs consist of a generator and discriminator trained adversarially. The generator improves at creating realistic data while the discriminator improves at detection. Used for image generation, style transfer, and data augmentation.
Examples
- StyleGAN for face generation
- CycleGAN for image-to-image translation
- Pix2Pix for paired image translation