Masked Autoregressive Flow for Density Estimation

George Papamakarios, Theo Pavlakou, Iain Murray

5/19/2017

stat.MLcs.LG

Official code: gpapamak/maf (225 stars, last commit 2024-07-28).

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.

View on arXiv View PDF

Code Implementations(1)

gpapamak/mafOfficial100%

Masked Autoregressive Flow

22542PythonMay 19, 20172 years agoNOASSERTION

Similar Papers

Neural Autoregressive Flows

86% match1 repo1252018

Block Neural Autoregressive Flow

85% match1 repo1832019

Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design

83% match1 repo1922019

Diffusion Density Estimators

83% match2024

Probabilistic Classification by Density Estimation Using Gaussian Mixture Model and Masked Autoregressive Flow

83% match2023

Kernelised Normalising Flows

83% match2023

Cite this paper

@article{papamakarios2017masked,
  title  = {Masked Autoregressive Flow for Density Estimation},
  author = {George Papamakarios and Theo Pavlakou and Iain Murray},
  year   = {2017},
  eprint = {1705.07057},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1705.07057}
}

Add a badge to your README

Shows whether official code is linked for this paper.

[![Code of Paper](https://codeofpaper.com/badge/1705.07057.svg)](https://codeofpaper.com/paper/1705.07057)