InfoGAN-CR and ModelCentrality: Self-supervised Model Training and Selection for Disentangling GANscs.LG
Improves disentangled GAN training and selection without labeled data.
problem Challenges in training disentangled GANs, especially self-supervision.
method Contrastive Regularizer and ModelCentrality for unsupervised disentanglement.
result Significantly improved disentanglement scores without labeled data.