AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Using the Gans Argument in Machine Learning?
Gans, generative eversarior network has been called a one of the coolest tricks in machine learning. The trick is basically yt to set up one network that would do thitw try to do this generative predictive coding, and another network that was basically to try to compete with the generator. So you take an image and say, where this is generated by the generator, or is it actually an esample from the world? And then the two can then compete. If the discriminator catches the generator a making a forgery, if it doesn't look enough like a, look enough likea, like a real word sample, then he would try to punish the generators