0
$\begingroup$

What is the application of Generative Adversarial Networks having been successfully trained?

splitted into two part as G and D, the G is for creation, and the D for a decider? Then there is a assumption that is the input of GAN have to be continued?

$\endgroup$
1
$\begingroup$

GANs were invented in a bar somewhere in Montreal, Canada. At said bar, the idea was that neural networks could be used for generating new examples from an existing distribution. This was the problem:

Given an input set X, can we make a new x’ that looks like it should be in X?

The classic description of a GAN is a counterfeiter (generator) and a cop (discriminator). The counterfeiter has the same problem, make a piece of paper look like real currency.

In training a GAN, the input to the generator is random noise, a starting seed so that no 2 results are the same. The generator then makes a new x’. The input to the discriminator alternates between an actual x and an x’ that the generator made. The discriminator then takes the and decides whether it is part of the set X. The discriminator is then trained using its answer to ensure that it can properly tell the difference between bad counterfeits and elements of X. When the discriminator makes a decision on an x’ that the generator made, the generator is updated as well, in order to increase its ability to make new x’s that the discriminator will think are in X.

Using this simple framework, these 2 networks works against each other (adversarially) to train each other until the generator is making x’s so well that the discriminator can’t tell the difference between them and the real thing. At this point, the generator can be used to make new pictures of cats or whatever was the goal in the first place.

$\endgroup$
  • $\begingroup$ Then there is no usage or application of the D after successful training? It is just for training the G in an unsupervised way? $\endgroup$ – XL _At_Here_There Jan 29 '18 at 13:16
  • $\begingroup$ It’s primary usage is training G in a (sorta) supervised way but after G is trained, it isn’t needed for generation. That’s not to say having a good discriminator isn’t useful. There are many applications where that’s exactly what you’d want but if the goal is generation then probably not. If this answers you question, you should accept this answer. $\endgroup$ – Jaden Travnik Jan 29 '18 at 13:35
  • $\begingroup$ Why do you think it is trained in supervised way? I think it would be unsupervised. $\endgroup$ – XL _At_Here_There Jan 29 '18 at 14:24
  • $\begingroup$ The generator is trained in a (sorta) supervised fashion because there is actually a label associated with the x' it generates (whether the discriminator could detect that it was a counterfeit or not). Unsupervised methods are used when there are no labels and one wants to understand how their data either by clustering or finding a good distance function. Refer to ai.stackexchange.com/questions/4706/… for to see the difference between supervised and unsupervised. $\endgroup$ – Jaden Travnik Jan 29 '18 at 15:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.