Calculus of Variations and Geometric Measure Theory

C. Finlay - A. Gerolin - Adam M Oberman - A. A. Pooladian

Learning normalizing flows from Entropy-Kantorovich potentials

created by gerolin on 23 Jun 2020



Inserted: 23 jun 2020

Year: 2020

ArXiv: 2006.06033 PDF


We approach the problem of learning continuous normalizing flows from a dual perspective motivated by entropy-regularized optimal transport, in which continuous normalizing flows are cast as gradients of scalar potential functions. This formulation allows us to train a dual objective comprised only of the scalar potential functions, and removes the burden of explicitly computing normalizing flows during training. After training, the normalizing flow is easily recovered from the potential functions.