mlsanityData to 14 Oct 2026

Flowification: Everything is a Normalizing Flow

arXiv 2205.15209 ↗First posted 30 May 2022cs.LG stat.MLcited by 5

Not on the map. Why?

Abstract (arXiv)

The two key characteristics of a normalizing flow is that it is invertible (in particular, dimension preserving) and that it monitors the amount by which it changes the likelihood of data points as samples are propagated along the network. Recently, multiple generalizations of normalizing flows have been introduced that relax these two conditions. On the other hand, neural networks only perform a forward pass on the input, there is neither a notion of an inverse of a neural network nor is there one of its likelihood contribution. In this paper we argue that certain neural network architectures can be enriched with a stochastic inverse pass and that their likelihood contribution can be monitored in a way that they fall under the generalized notion of a normalizing flow mentioned above. We term this enrichment flowification. We prove that neural networks only containing linear layers, convolutional layers and invertible activations such as LeakyReLU can be flowified and evaluate them in the generative setting on image datasets.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 25 of its 42 references that are in the map. First posted on arXiv 30 May 2022.

Builds on

25 of 42 references.

References · 25

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 5,045
  4. 2016Density estimation using Real NVPcited by 2,042
  5. 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,545
  6. 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,323
  7. 2017Masked Autoregressive Flow for Density Estimationcited by 684
  8. 2016Improving Variational Inference with Inverse Autoregressive Flowcited by 640
  9. 2019Neural Spline Flowscited by 596
  10. 2018Do Deep Generative Models Know What They Don't Know?cited by 443
  11. 2018Invertible Residual Networkscited by 365
  12. 2015A note on the evaluation of generative modelscited by 292
13 more
  1. 2019Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Designcited by 262
  2. 2019Residual Flows for Invertible Generative Modelingcited by 257
  3. 2020Flows for simultaneous manifold learning and density estimationcited by 118
  4. 2020SurVAE Flows: Surjections to Bridge the Gap between VAEs and Flowscited by 71
  5. 2013RNADE: The real-valued neural autoregressive density-estimatorcited by 66
  6. 2020Augmented Normalizing Flows: Bridging the Gap Between Generative Flows and Latent Variable Modelscited by 66
  7. 2021Rectangular Flows for Manifold Learningcited by 50
  8. 2021CaloFlow: Fast and Accurate Generation of Calorimeter Showers with Normalizing Flowscited by 46
  9. 2016Improving Variational Auto-Encoders using Householder Flowcited by 42
  10. 2019Emerging Convolutions for Generative Normalizing Flowscited by 31
  11. 2021Tractable Density Estimation on Learned Manifolds with Conformal Embedding Flowscited by 31
  12. 2020The Convolution Exponential and Generalized Sylvester Flowscited by 18
  13. 2021Funnels: Exact maximum likelihood with dimensionality reductioncited by 4

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 0; 2025: 2; 2026: 2.

5 papers.

  1. 2023Learning Interpolations between Boltzmann Densitiescited by 43
  2. 2025FARMER: Flow AutoRegressive Transformer over Pixelscited by 18
  3. 2025SimFlow: Simplified and End-to-End Training of Latent Normalizing Flowscited by 7
  4. 2026Learning Energy-Based Models from Stochastic Interpolants using Spatiotemporal Differencescited by 2
  5. 2026Learning to Align Generative Appearance Priors for Fine-grained Image Retrievalcited by 0