Masked Autoregressive Flow for Density Estimation

TL;DR

This paper introduces Masked Autoregressive Flow, a flexible normalizing flow model that enhances autoregressive density estimators by stacking models to improve performance, achieving state-of-the-art results.

Contribution

It proposes a novel stacking approach for autoregressive models to create Masked Autoregressive Flow, generalizing existing flows like Real NVP and Inverse Autoregressive Flow.

Findings

Achieves state-of-the-art density estimation performance

Demonstrates flexibility and effectiveness of the stacking approach

Generalizes previous normalizing flow models

Abstract

Autoregressive models are among the best performing neural density estimators. We describe an approach for increasing the flexibility of an autoregressive model, based on modelling the random numbers that the model uses internally when generating data. By constructing a stack of autoregressive models, each modelling the random numbers of the next model in the stack, we obtain a type of normalizing flow suitable for density estimation, which we call Masked Autoregressive Flow. This type of flow is closely related to Inverse Autoregressive Flow and is a generalization of Real NVP. Masked Autoregressive Flow achieves state-of-the-art performance in a range of general-purpose density estimation tasks.