Jonathan Ho

Jonathan Ho is a machine learning researcher best known as the lead author of "Denoising Diffusion Probabilistic Models" (DDPM), the 2020 paper that made diffusion models practical for high quality image generation and launched the modern diffusion based text-to-image wave. He is also credited with classifier-free guidance, cascaded diffusion models, and video diffusion models, and was a coauthor on Google's Imagen. Ho earned his PhD at the University of California, Berkeley, under Pieter Abbeel, worked at OpenAI and Google Brain, and in 2022 cofounded the image generation company Ideogram. ^{[1][2][3][19]}

The DDPM paper is among the most cited works in modern generative modeling: as of 2026 it had accumulated roughly 42,000 citations on Google Scholar, and Ho's overall citation count exceeds 97,000 with an h-index of 40. ^[18]

Beyond DDPM, the techniques most directly associated with Ho, the DDPM noise prediction objective and classifier-free guidance, appear in nearly every well known diffusion image generator, including OpenAI's DALL-E 2, Google's Imagen, and the open source Stable Diffusion family. ^[4][5][6][7]

Who is Jonathan Ho?

Jonathan Ho is a machine learning researcher whose work sits at the center of the shift from generative adversarial networks to diffusion models as the dominant approach to image synthesis. His research spans imitation learning, normalizing flows, and the diffusion models he became known for. He completed his PhD in the Electrical Engineering and Computer Sciences department at the University of California, Berkeley, in 2020, advised by Pieter Abbeel, and his dissertation, "Deep Generative Models: Imitation Learning, Image Synthesis, and Compression," gathered his work on imitation learning, flow based models, and diffusion. ^[1][9]

His first widely cited result came in 2016, before he finished his doctorate, in a collaboration with Stefano Ermon on "Generative Adversarial Imitation Learning" (GAIL). That paper, presented at NeurIPS 2016, framed imitation learning as a problem that could be solved with the adversarial training ideas used in generative adversarial networks, and it produced a model free algorithm that learned to copy expert behavior in high dimensional control tasks. The work was carried out while Ho was associated with Stanford and OpenAI, where he was a researcher around 2016 to 2017, and Ermon was an early collaborator who helped shape his graduate research. GAIL has itself been cited more than 5,000 times. ^[8][18]

During this period he also worked on likelihood based generative models. In 2019 he was the lead author of "Flow++," presented at ICML, which improved flow based generative models through variational dequantization and a redesigned architecture. The coauthors included Xi Chen, Aravind Srinivas, Yan Duan, and Pieter Abbeel. Flow++ pushed normalizing flows toward stronger density estimation results, and the experience with likelihood based image models fed directly into his later diffusion work. ^[10]

What is DDPM?

Diffusion probabilistic models had been proposed earlier, most notably by Jascha Sohl-Dickstein and colleagues in 2015, but they had not yet produced image samples competitive with the best generative adversarial networks. The idea is to define a forward process that gradually adds Gaussian noise to an image until it becomes pure noise, then to train a neural network to reverse that process step by step, turning noise back into a sample from the data distribution. ^[2][11]

In "Denoising Diffusion Probabilistic Models," posted to arXiv in June 2020 and presented at NeurIPS 2020, Ho, Ajay Jain, and Pieter Abbeel rewrote the training objective in a form that was simple to optimize. Rather than predicting the reverse distribution directly, the network was trained to predict the noise that had been added at each step, using a weighted variant of the variational bound that reduced to a plain mean squared error loss. The authors connected this objective to denoising score matching and to Langevin dynamics, which gave the method a clear theoretical footing. The paper states the central result directly: "Our best results are obtained by training on a weighted variational bound designed according to a novel connection between diffusion probabilistic models and denoising score matching with Langevin dynamics." On the unconditional CIFAR-10 benchmark the model reached an Inception score of 9.46 and a Frechet Inception Distance of 3.17, results that were state of the art for the dataset at the time and that put diffusion models on the same level as the leading GANs. ^[2][11]

The paper mattered because it made diffusion models practical. The training procedure was stable and did not require the adversarial balancing act that made GANs hard to train, and the loss was easy to implement. Within a short time the approach was extended, scaled, and adopted across the field, and DDPM became one of the most cited papers in modern generative modeling, with roughly 42,000 citations as of 2026. ^[2][3][18]

How does DDPM relate to score-based models?

The rise of diffusion models is often told together with the parallel line of score-based generative models developed by Yang Song and Stefano Ermon. Their 2019 work on noise conditional score networks estimated the gradient of the data density, called the score, at multiple noise levels and used Langevin dynamics to sample. The two approaches turned out to be closely related. In 2021 Song and coauthors, including Ermon, framed both DDPM and score-based models as discretizations of a single continuous time process described by a stochastic differential equation, which unified the two views and clarified why the denoising objective in DDPM worked. Ho's contribution sat on the diffusion side of this convergence, and the noise prediction objective he introduced is one of the standard ways the score is learned in practice. ^[12][13]

What else has Jonathan Ho contributed to diffusion models?

After his PhD, Ho continued to develop the diffusion framework at Google, and several of his later methods became default components of large generative systems.

In 2022 he and Tim Salimans introduced classifier-free guidance in the paper "Classifier-Free Diffusion Guidance." Earlier work had improved sample quality by steering a diffusion model with the gradient of a separate image classifier, an approach called classifier guidance. Ho and Salimans showed that the same trade off between sample quality and diversity could be reached without any extra classifier. Their method trains one network to act as both a conditional and an unconditional model, then combines the two score estimates at sampling time using a single guidance weight. The technique was simple to apply and became the standard way to control text-to-image diffusion models, used in systems such as DALL-E 2, GLIDE, and Imagen. The paper has been cited more than 8,000 times. ^[4][14][18]

Ho also worked on building diffusion models that generate high resolution images. In "Cascaded Diffusion Models for High Fidelity Image Generation," published in the Journal of Machine Learning Research, he and coauthors including Chitwan Saharia, William Chan, David Fleet, Mohammad Norouzi, and Tim Salimans chained several diffusion models together. A base model produces a small image, and a sequence of super-resolution diffusion models upsamples it to higher resolutions. A method they called conditioning augmentation, which adds noise to the low resolution inputs during training, was important for keeping sample quality high through the cascade. The pipeline produced class-conditional ImageNet samples that outperformed strong GAN and autoregressive baselines on Frechet Inception Distance. ^[5]

These ideas came together in Imagen, Google's text-to-image system described in the 2022 paper "Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding." Ho was one of the coauthors. Imagen paired a large frozen text encoder, taken from a pretrained language model, with a cascade of diffusion models and classifier-free guidance, and it reported strong results on photorealism and image text alignment. The system drew directly on the guidance and cascade techniques that Ho had helped develop, and the Imagen paper has itself been cited nearly 10,000 times. ^[6][18]

Ho then extended diffusion to moving images. In "Video Diffusion Models," presented in 2022, he and coauthors adapted the image diffusion architecture to space and time so that it could generate short video clips, an early demonstration that the same training recipe could handle temporal data. This line of work led to Imagen Video, a text-conditional video generation system built from a cascade of video diffusion models with classifier-free guidance, which Ho coauthored later in 2022. ^[7][15]

How did Jonathan Ho influence text-to-image generation?

The denoising diffusion framework that Ho helped establish became the technical foundation of the text-to-image boom that began around 2022. The combination of a diffusion model and classifier-free guidance underlies most of the well known image generators of that period, including OpenAI's DALL-E 2, Google's Imagen, and the open source Stable Diffusion family. The two methods most directly associated with Ho, the DDPM training objective and classifier-free guidance, appear in nearly all of these systems. ^[3][14]

Diffusion models also displaced generative adversarial networks as the dominant approach to high quality image synthesis in research, in part because they were easier to train and scaled well with data and compute. The reach of the work extended past still images into video, audio, and other domains, and the noise prediction recipe from the 2020 paper remains a common starting point for new diffusion systems. ^[3][12]

Where has Jonathan Ho worked?

Ho's career has moved through several of the leading machine learning research organizations. He was a researcher at OpenAI around 2016 to 2017, the period of the GAIL work, before and during the start of his Berkeley PhD. After completing his doctorate in 2020, he joined Google as a research scientist, working at Google Brain on the diffusion and text-to-image projects described above. Google Brain was later merged into Google DeepMind in 2023. ^[1][6][8]

In 2022 Ho cofounded Ideogram, a company focused on text-to-image generation, together with Mohammad Norouzi, William Chan, and Chitwan Saharia, all former Google Brain researchers who had worked on related generative modeling and Imagen research. The company is based in Toronto. It came out of stealth in August 2023 with an initial model and announced 16.5 million dollars in seed funding led by Andreessen Horowitz and Index Ventures. Ideogram drew attention for generating legible text inside images, a task that earlier image generators had struggled with, and it raised an 80 million dollar Series A round in February 2024 alongside the release of its 1.0 model. ^[3][16][17]

Ho is listed as a cofounder of Ideogram, but professional profiles indicate he has since left the company, which is led by Mohammad Norouzi as chief executive. As of 2026 those profiles list Ho as a research scientist at Meta. This current affiliation is reported by professional networking and contact aggregators rather than a primary statement from Ho, and his personal website was not publishing an updated biography at the time of writing. ^[18][19]

What recognition has Jonathan Ho received?

Ho's standing in the field rests mainly on the influence of his research rather than on formal prizes. "Denoising Diffusion Probabilistic Models" is among the most cited papers in generative modeling, with roughly 42,000 citations, and classifier-free guidance and cascaded diffusion models are widely used building blocks named after the techniques he introduced. By 2026 his combined work had been cited more than 97,000 times, with an h-index of 40. Commentary on the founders of Ideogram routinely describes him as the lead author of the DDPM paper that defined the modern diffusion framework used by consumer image generators. The score-based modeling work that converged with his diffusion research, led by Yang Song and Stefano Ermon, received an Outstanding Paper Award at ICLR 2021, which reflects the broader recognition the surrounding area attracted. ^{[2][3][13][18]}

Facts

Scroll sideways for more →

References

1. Ho, Jonathan. "Deep Generative Models: Imitation Learning, Image Synthesis, and Compression." PhD dissertation, EECS Department, University of California, Berkeley, 2020. https://www2.eecs.berkeley.edu/Pubs/TechRpts/2020/EECS-2020-67.html ↩
2. Ho, Jonathan, Jain, Ajay, and Abbeel, Pieter. "Denoising Diffusion Probabilistic Models." arXiv:2006.11239, 2020. https://arxiv.org/abs/2006.11239 ↩
3. "Ideogram (text-to-image model)." Wikipedia, accessed 2026. https://en.wikipedia.org/wiki/Ideogram_(text-to-image_model) ↩
4. Ho, Jonathan, and Salimans, Tim. "Classifier-Free Diffusion Guidance." arXiv:2207.12598, 2022. https://arxiv.org/abs/2207.12598 ↩
5. Ho, Jonathan, Saharia, Chitwan, Chan, William, Fleet, David J., Norouzi, Mohammad, and Salimans, Tim. "Cascaded Diffusion Models for High Fidelity Image Generation." Journal of Machine Learning Research, 2022. https://arxiv.org/abs/2106.15282 ↩
6. Saharia, Chitwan, et al. "Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding" (Imagen). arXiv:2205.11487, 2022. https://arxiv.org/abs/2205.11487 ↩
7. Ho, Jonathan, Salimans, Tim, Gritsenko, Alexey, Chan, William, Norouzi, Mohammad, and Fleet, David J. "Video Diffusion Models." arXiv:2204.03458, 2022. https://arxiv.org/abs/2204.03458 ↩
8. Ho, Jonathan, and Ermon, Stefano. "Generative Adversarial Imitation Learning." Advances in Neural Information Processing Systems 29 (NeurIPS), 2016. https://arxiv.org/abs/1606.03476 ↩
9. "Pieter Abbeel." Berkeley Research, University of California, Berkeley, accessed 2026. https://vcresearch.berkeley.edu/faculty/pieter-abbeel ↩
10. Ho, Jonathan, Chen, Xi, Srinivas, Aravind, Duan, Yan, and Abbeel, Pieter. "Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design." Proceedings of ICML, 2019. https://arxiv.org/abs/1902.00275 ↩
11. Ho, Jonathan, Jain, Ajay, and Abbeel, Pieter. "Denoising Diffusion Probabilistic Models" (project page). 2020. https://hojonathanho.github.io/diffusion/ ↩
12. Song, Yang, and Ermon, Stefano. "Generative Modeling by Estimating Gradients of the Data Distribution." Advances in Neural Information Processing Systems 32 (NeurIPS), 2019. https://arxiv.org/abs/1907.05600 ↩
13. Song, Yang, Sohl-Dickstein, Jascha, Kingma, Diederik P., Kumar, Abhishek, Ermon, Stefano, and Poole, Ben. "Score-Based Generative Modeling through Stochastic Differential Equations." ICLR 2021 (Outstanding Paper). https://arxiv.org/abs/2011.13456 ↩
14. "Imagen Video: High Definition Video Generation with Diffusion Models." arXiv:2210.02303, 2022. https://arxiv.org/abs/2210.02303 ↩
15. Ho, Jonathan, et al. "Video Diffusion Models" (project page). 2022. https://video-diffusion.github.io/ ↩
16. Norouzi, Mohammad. "Announcing the formation of Ideogram." Ideogram, 2023. https://ideogram.ai/launch ↩
17. "Pieter Abbeel: research group and alumni." People at EECS, University of California, Berkeley, accessed 2026. https://people.eecs.berkeley.edu/~pabbeel/group.html ↩
18. "Jonathan Ho." Google Scholar profile, accessed 2026. https://scholar.google.com/citations?user=iVLAQysAAAAJ ↩
19. "Jonathan Ho, Research Scientist." Professional profile listing, accessed 2026. https://www.linkedin.com/in/jonathan-ho-00243623/ ↩