Log In Sign Up

Multi-Resolution Continuous Normalizing Flows

by   Vikram Voleti, et al.

Recent work has shown that Neural Ordinary Differential Equations (ODEs) can serve as generative models of images using the perspective of Continuous Normalizing Flows (CNFs). Such models offer exact likelihood calculation, and invertible generation/density estimation. In this work we introduce a Multi-Resolution variant of such models (MRCNF), by characterizing the conditional distribution over the additional information required to generate a fine image that is consistent with the coarse image. 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. Further, we examine the out-of-distribution properties of (Multi-Resolution) Continuous Normalizing Flows, and find that they are similar to those of other likelihood-based generative models.


page 8

page 16


FFJORD: Free-form Continuous Dynamics for Scalable Reversible Generative Models

A promising class of generative models maps points from a simple distrib...

MaCow: Masked Convolutional Generative Flow

Flow-based generative models, conceptually attractive due to tractabilit...

Equivariant Normalizing Flows for Point Processes and Sets

A point process describes how random sets of exchangeable points are gen...

TzK: Flow-Based Conditional Generative Model

We formulate a new class of conditional generative models based on proba...

InfoCNF: An Efficient Conditional Continuous Normalizing Flow with Adaptive Solvers

Continuous Normalizing Flows (CNFs) have emerged as promising deep gener...

Stochastic Normalizing Flows

We introduce stochastic normalizing flows, an extension of continuous no...

Granular Learning with Deep Generative Models using Highly Contaminated Data

An approach to utilize recent advances in deep generative models for ano...