- Q1: Deep Convolutional GAN (DCGAN)
- Papers:
- Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
- One-sided Label Smoothing (Improved Techniques for Training GANs)
- Add Noise (Towards Principled Methods for Training Generative Adversarial Networks)
- Q2: AC-GAN, WGAN
- Papers:
- Conditional Image Synthesis With Auxiliary Classifier GANs
- Wasserstein GAN (Arjovsky et al. 2017)