mlsanityData to 7 Oct 2026

A Review of Change of Variable Formulas for Generative Modeling

arXiv 2308.02652 ↗First posted 4 Aug 2023cs.LGcited by 12

Not on the map. Why?

Abstract (arXiv)

Change-of-variables (CoV) formulas allow to reduce complicated probability densities to simpler ones by a learned transformation with tractable Jacobian determinant. They are thus powerful tools for maximum-likelihood learning, Bayesian inference, outlier detection, model selection, etc. CoV formulas have been derived for a large variety of model types, but this information is scattered over many separate works. We present a systematic treatment from the unifying perspective of encoder/decoder architectures, which collects 28 CoV formulas in a single place, reveals interesting relationships between seemingly diverse methods, emphasizes important distinctions that are not always clear in the literature, and identifies surprising gaps for future research.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 83 of its 111 references that are in the map. First posted on arXiv 4 Aug 2023.

Builds on

83 of 111 references.

References · 83

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  4. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  5. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  6. 2022Flow Matching for Generative Modelingcited by 5,426
  7. 2017Neural Discrete Representation Learningcited by 4,650
  8. 2018Neural Ordinary Differential Equationscited by 3,733
  9. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,550
  10. 2015Variational Inference with Normalizing Flowscited by 2,161
  11. 2016Density estimation using Real NVPcited by 2,029
  12. 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,539
38 more
  1. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,376
  2. 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,315
  3. 2014NICE: Non-linear Independent Components Estimationcited by 1,052
  4. 2016Conditional Image Generation with PixelCNN Decoderscited by 859
  5. 2017Masked Autoregressive Flow for Density Estimationcited by 680
  6. 2021Score-based Generative Modeling in Latent Spacecited by 634
  7. 2019Neural Spline Flowscited by 590
  8. 2018FFJORD: Free-form Continuous Dynamics for Scalable Reversible Generative Modelscited by 584
  9. 2019The frontier of simulation-based inferencecited by 490
  10. 2018Invertible Residual Networkscited by 364
  11. 2019Residual Flows for Invertible Generative Modelingcited by 256
  12. 2013Causal Discovery with Continuous Additive Noise Modelscited by 252
  13. 2019DiffTaichi: Differentiable Programming for Physical Simulationcited by 248
  14. 2018Neural Autoregressive Flowscited by 207
  15. 2019Rethinking Lossy Compression: The Rate-Distortion-Perception Tradeoffcited by 207
  16. 2016Variational Lossy Autoencodercited by 175
  17. 2019Learning Likelihoods with Conditional Normalizing Flowscited by 170
  18. 2020Stochastic Normalizing Flowscited by 156
  19. 2018Neural Importance Samplingcited by 154
  20. 2020Riemannian Continuous Normalizing Flowscited by 143
  21. 2019From Variational to Deterministic Autoencoderscited by 131
  22. 2020Normalizing Flows on Tori and Spherescited by 117
  23. 2020Flows for simultaneous manifold learning and density estimationcited by 117
  24. 2020Learning Latent Space Energy-Based Prior Modelcited by 104
  25. 2020SoftFlow: Probabilistic Framework for Normalizing Flow on Manifoldscited by 97
  26. 2020Causal Autoregressive Flowscited by 97
  27. 2021The Causal-Neural Connection: Expressiveness, Learnability, and Inferencecited by 97
  28. 2021Independent mechanism analysis, a new concept?cited by 96
  29. 2020Invertible Image Rescalingcited by 92
  30. 2020Disentanglement by Nonlinear ICA with General Incompressible-flow Networks (GIN)cited by 91
  31. 2022Multi-Realism Image Compression with a Conditional Generatorcited by 79
  32. 2019Semi-Supervised Learning with Normalizing Flowscited by 76
  33. 2020SurVAE Flows: Surjections to Bridge the Gap between VAEs and Flowscited by 71
  34. 2016Normalizing Flows on Riemannian Manifoldscited by 68
  35. 2020Augmented Normalizing Flows: Bridging the Gap Between Generative Flows and Latent Variable Modelscited by 66
  36. 2019Sum-of-Squares Polynomial Flowcited by 62
  37. 2018Deep Generative Models for Distribution-Preserving Lossy Compressioncited by 59
  38. 2020A Disentangling Invertible Interpretation Network for Explaining Latent Representationscited by 57

50 most cited of 83 All 83 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 5; 2025: 3; 2026: 2.

12 papers.

  1. 2024Deep Generative Models through the Lens of the Manifold Hypothesis: A Survey and New Connectionscited by 47
  2. 2023Consistency Models for Scalable and Fast Simulation-Based Inferencecited by 22
  3. 2024On the Universality of Volume-Preserving and Coupling-Based Normalizing Flowscited by 18
  4. 2023Leveraging Self-Consistency for Data-Efficient Amortized Bayesian Inferencecited by 14
  5. 2025Robust Amortized Bayesian Inference with Self-Consistency Losses on Unlabeled Datacited by 11
  6. 2025Mitigating Diffusion Model Hallucinations with Dynamic Guidancecited by 5
  7. 2024Analyzing Generative Models by Manifold Entropic Metricscited by 2
  8. 2026Physics-informed neural particle flow for the Bayesian update stepcited by 2
  9. 2024Differentiable Score-Based Likelihoods: Learning CT Motion Compensation From Clean Imagescited by 1
  10. 2024Score Change of Variablescited by 0
  11. 2025Time-dependent density estimation using binary classifierscited by 0
  12. 2026Unsupervised Continual Learning for Amortized Bayesian Inferencecited by 0