Nonlinear Isometric Manifold Learning for Injective Normalizing Flows

03/08/2022

∙

by   Eike Cramer, et al.

∙

0

∙

share

To model manifold data using normalizing flows, we propose to employ the isometric autoencoder to design nonlinear encodings with explicit inverses. The isometry allows us to separate manifold learning and density estimation and train both parts to high accuracy. Applied to the MNIST data set, the combined approach generates high-quality images.