Improving Continuous Normalizing Flows using a Multi-Resolution Framework
2021-06-02ICML Workshop INNF 2021Unverified0· sign in to hype
Vikram Voleti, Chris Finlay, Adam M Oberman, Christopher Pal
Unverified — Be the first to reproduce this paper.
ReproduceAbstract
Recent work has shown that Continuous Normalizing Flows (CNFs) can serve as generative models of images with exact likelihood calculation and invertible generation/density estimation. In this work we introduce a Multi-Resolution variant of such models (MRCNF). We introduce a transformation between resolutions that allows for no change in the log likelihood. We show that this approach yields comparable likelihood values for various image datasets, with improved performance at higher resolutions, with fewer parameters, using only 1 GPU.