Variational GAN

[Yellowwww]

I have read your paper recently ,in this paper, you use the variational conditional GAN to approximate robust probability distributions for the task of approximating accurate conditional channel distribution p(y|x) ,in the structure of your network, you use an estimation network and a discriminative network to accomplish this work, I have a question about how you get the final estimation of p(y|x),did you input a batch of x into the generator to produce a batch of Y, and you use the Monte Carlo Simulation to get the final estimated conditional channel distribution p(y|x). Thank you very much if you could give me a reply.