As mentioned, there are new papers on GANs coming out every month (as you can check on the reference table made by Hindu Puravinash that we mentioned at the beginning of the chapter).
Anyway, apart from the vanilla implementation described in the initial paper from Goodfellow and his colleagues, the most notable implementations to take notice of are deep convolutional generative adversarial networks (DCGANs) and conditional GANs (CGANs).
- DCGANs are GANs based on CNN architecture (
RADFORD, Alec; METZ, Luke; CHINTALA, Soumith. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:1511.06434
, 2015
: https://arxiv.org/abs/1511.06434). - CGANs are DCGANs which are ...