Learning disconnected manifolds: a no GANs land
arXiv:2006.04596
Abstract
Typical architectures of Generative AdversarialNetworks make use of a unimodal latent distribution transformed by a continuous generator. Consequently, the modeled distribution always has connected support which is cumbersome when learning a disconnected set of manifolds. We formalize this problem by establishing a no free lunch theorem for the disconnected manifold learning stating an upper bound on the precision of the targeted distribution. This is done by building on the necessary existence of a low-quality region where the generator continuously samples data between two disconnected modes. Finally, we derive a rejection sampling method based on the norm of generators Jacobian and show its efficiency on several generators including BigGAN.
arXiv comment24 pages
References in corpus (14)
- Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- Large Scale GAN Training for High Fidelity Natural Image Synthesis
- Spectral Normalization for Generative Adversarial Networks
- Self-Attention Generative Adversarial Networks
- VEEGAN: Reducing Mode Collapse in GANs using Implicit Variational Learning
- Assessing Generative Models via Precision and Recall
- AdaGAN: Boosting Generative Models
- On the regularization of Wasserstein GANs
- Disconnected Manifold Learning for Generative Adversarial Networks
- Discriminator Rejection Sampling
- Some Theoretical Properties of GANs
- MMGAN: Generative Adversarial Networks for Multi-Modal Distributions
- Rethinking Generative Mode Coverage: A Pointwise Guaranteed Approach