I'm working in a binary classification problem. In general terms, I'm focused on using neural networks to classify datasets that are not huge and in order to address that problem I'm comparing different data augmentation methods.
One of these methods is using a generative network (e.g. an MLP whose output is a d-dimensional vector) trained in an adversarial way. My question is, since I have two classes: is a good (even, possible) idea to use two generators? (one per class) If it's not, how can I make sure which class is generating when using generator's forward pass when training is over?
Thanks in advance. I'd really appreciate references (such as texts or even papers) and any advice ^^
