Yann LeCun, Director of Facebook AI, has recently stated that "Generative Adversarial Networks is the most interesting idea in the last ten years in machine learning", and that is certainly confirmed by the elevated interest in academia about this deep learning solution. If you look at recent papers on deep learning (but also look at the leading trends on LinkedIn or Medium posts on the topic), there has really been an overproduction of variants of GANs.
You can get an idea of what a zoo the world of GANs has become just by glancing the continuously updated reference table, created by Hindu Puravinash, which can be found at https://github.com/hindupuravinash/the-gan-zoo/blob/master/gans.tsv or ...
