|
|
|
|
|
|
by cjalmeida
3116 days ago
|
|
|
The "novelty" behind "Generative Adversarial Networks" (GANs), a very popular deep learning network, is based on Game Theory. It's a "game" between a Generator and a Discriminator networks where one try to fool the other. The Generator tries to fool the Discriminator by producing output the Discriminator can't distinguish between fake or real. You can find real-word applications for GANs (specially the Pix2Pix variety) in a lot of domains. |
|
|