Quantifying Manifolds: Do the manifolds learned by Generative Adversarial Networks converge to the real data manifold
This addresses the problem of understanding GAN convergence for researchers, though it appears incremental as it focuses on experimental quantification rather than new methods.
The paper investigates whether the manifolds learned by Generative Adversarial Networks (GANs) converge to the real data manifold during training by quantifying intrinsic dimensions and topological features, finding that these metrics approach those of the real data as training progresses.
This paper presents our experiments to quantify the manifolds learned by ML models (in our experiment, we use a GAN model) as they train. We compare the manifolds learned at each epoch to the real manifolds representing the real data. To quantify a manifold, we study the intrinsic dimensions and topological features of the manifold learned by the ML model, how these metrics change as we continue to train the model, and whether these metrics convergence over the course of training to the metrics of the real data manifold.