Normalizing flows are diffeomorphic, typically dimension-preserving, models trained using the likelihood of the model. We use the SurVAE framework to construct dimension reducing surjective flows via a new layer, known as the funnel. We demonstrate its efficacy on a variety of datasets, and show it improves upon or matches the performance of existing flows while having a reduced latent space size. The funnel layer can be constructed from a wide range of transformations including restricted convolution and feed forward layers.
@article{arxiv.2112.08069,
title = {Funnels: Exact maximum likelihood with dimensionality reduction},
author = {Samuel Klein and John A. Raine and Sebastian Pina-Otey and Slava Voloshynovskiy and Tobias Golling},
journal= {arXiv preprint arXiv:2112.08069},
year = {2021}
}