HIGHLIGHTS
SUMMARY
The authors present IberianGAN, a framework based on generative models that reconstruct pottery profiles from rim or base fragments (see Fig 1A,B).. A typical GAN33 framework contains a generative (G) and a discriminative (D) neural_network such that G aims to generate realistic samples, while D learns to discriminate if a sample is from the real data distribution (H0) or not. D tries to maximize (log D(x)), which is the probability of having a correct classification of actual shapes, while G tries to minimize (log (1 - D(G(x))), which is the probability . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.