Improving Continuous Normalizing Flows Using a Multi-Resolution Framework
Abstract
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.
Cite
Text
Voleti et al. "Improving Continuous Normalizing Flows Using a Multi-Resolution Framework." ICML 2021 Workshops: INNF, 2021.Markdown
[Voleti et al. "Improving Continuous Normalizing Flows Using a Multi-Resolution Framework." ICML 2021 Workshops: INNF, 2021.](https://mlanthology.org/icmlw/2021/voleti2021icmlw-improving/)BibTeX
@inproceedings{voleti2021icmlw-improving,
title = {{Improving Continuous Normalizing Flows Using a Multi-Resolution Framework}},
author = {Voleti, Vikram and Finlay, Chris and Oberman, Adam M and Pal, Christopher},
booktitle = {ICML 2021 Workshops: INNF},
year = {2021},
url = {https://mlanthology.org/icmlw/2021/voleti2021icmlw-improving/}
}