AI People Wiki

Builders & open source

Jonathan Ho

First author of Denoising Diffusion Probabilistic Models (DDPM) and of Classifier-Free Diffusion Guidance

Jonathan Ho is first author of "Denoising Diffusion Probabilistic Models" (DDPM, NeurIPS 2020), written with Ajay Jain and Pieter Abbeel, all three listed with UC Berkeley on the paper's project page. With Tim Salimans, Ho wrote "Classifier-Free Diffusion Guidance" (2022). Ho is a co-first author of Imagen Video, a text-to-video system of cascaded video diffusion models presented by Google Research's Brain Team, and a co-author of Imagen and of "Variational Diffusion Models" with Diederik P. Kingma, Tim Salimans, and Ben Poole.1234567

At a glance

Current role
None confirmed in the sources
Education
UC Berkeley
Work listed
3
Sources cited
11
Profile reviewed

Research

Ho's early generative modeling papers were written with Pieter Abbeel, with DDPM listing its authors at UC Berkeley. Flow++ (2019) improved flow-based generative models with variational dequantization and new architecture choices, and "Denoising Diffusion Probabilistic Models" (2020), with Ajay Jain and Abbeel, showed high-quality image synthesis with diffusion models and drew connections to denoising score matching and Langevin dynamics.821

Ho's later papers, several written with Tim Salimans, extended diffusion models to higher resolutions and new modalities. Cascaded Diffusion Models (2021) chained diffusion models to generate high-fidelity images, "Video Diffusion Models" (2022) applied the approach to video, and "Classifier-Free Diffusion Guidance" (2022) showed that conditional and unconditional diffusion models can be combined to steer generation without a separate classifier.9104

Text-to-image and video systems

Ho co-authored Imagen (2022), a text-to-image diffusion model that uses a large pretrained language model to encode text for image synthesis, and is listed first among the equal-contribution authors of Imagen Video, which Google Research's Brain Team presented as a cascade of video diffusion models for text-conditional video generation.65

Work and contributions

  • Denoising Diffusion Probabilistic Models (DDPM) Published research

    First author, with Ajay Jain and Pieter Abbeel (all listed with UC Berkeley on the project page), of the NeurIPS 2020 paper on high-quality image synthesis with diffusion probabilistic models; the project page links the paper and the authors' code repository.213

  • Classifier-Free Diffusion Guidance Published research

    First author, with Tim Salimans, of the 2022 paper introducing classifier-free guidance for diffusion models.4

  • Imagen Video Published research

    Listed first among the equal-contribution authors of Imagen Video on Google Research's project page, which describes a text-conditional video generation system built from a cascade of video diffusion models.5

Timeline

  1. Posted Flow++, on improving flow-based generative models, as first author.8
  2. Published DDPM in the NeurIPS 2020 proceedings.3
  3. Posted "Denoising Diffusion Probabilistic Models" with Ajay Jain and Pieter Abbeel.21
  4. Posted "Cascaded Diffusion Models for High Fidelity Image Generation" as first author.9
  5. Co-authored "Variational Diffusion Models" with Diederik P. Kingma, Tim Salimans and Ben Poole.7
  6. Posted "Video Diffusion Models" as first author.10
  7. Co-authored the Imagen text-to-image diffusion paper.6
  8. Posted "Classifier-Free Diffusion Guidance" with Tim Salimans.4

Documented roles

The sources do not confirm a current role; these are the roles they document.

  • Listed with Google Research, Brain Team on Classifier-Free Diffusion Guidance and as co-first author of Imagen Video (2022), Google Research, Brain Team511
  • Listed with UC Berkeley as first author of DDPM (2020), UC Berkeley1

Companies in their history

Suggest a correction or removal

Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.

References

11 sources · profile reviewed Oct 21, 2026 · data updated Oct 21, 2026

  1. Denoising Diffusion Probabilistic Models — project pageJonathan Ho · Official page · accessed Oct 21, 2026
  2. Denoising Diffusion Probabilistic Models (arXiv 2006.11239)arXiv · Paper · published Jun 19, 2020 · accessed Oct 21, 2026
  3. Denoising Diffusion Probabilistic Models — NeurIPS 2020 proceedingsNeurIPS Proceedings · Paper · published 2020 · accessed Oct 13, 2026
  4. Classifier-Free Diffusion Guidance (arXiv 2207.12598)arXiv · Paper · published Jul 26, 2022 · accessed Oct 21, 2026
  5. Imagen Video — Google Research, Brain TeamGoogle Research · Official page · accessed Oct 21, 2026
  6. Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding (arXiv 2205.11487)arXiv · Paper · published May 23, 2022 · accessed Oct 21, 2026
  7. Variational Diffusion Models (arXiv 2107.00630)arXiv · Paper · published Jul 1, 2021 · accessed Oct 13, 2026
  8. Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design (arXiv 1902.00275)arXiv · Paper · published Feb 1, 2019 · accessed Oct 21, 2026
  9. Cascaded Diffusion Models for High Fidelity Image Generation (arXiv 2106.15282)arXiv · Paper · published May 30, 2021 · accessed Oct 21, 2026
  10. Video Diffusion Models (arXiv 2204.03458)arXiv · Paper · published Apr 7, 2022 · accessed Oct 21, 2026
  11. Classifier-Free Diffusion Guidance (arXiv PDF: author affiliations, Google Research, Brain team)arXiv · Paper · published Jul 26, 2022 · accessed Oct 13, 2026