diffusion model

American  

[dih-fyoo-zhuhn mod-l] / dɪˈfju ʒən ˌmɒd l /

noun

Computers.

1. a type of machine learning algorithm trained by having the algorithm add noise to images and then remove it from them: used especially for generating pictures from text prompts or for repairing, transforming, or expanding existing images.

---

Etymology

Origin of diffusion model

First recorded in 1995–2000

Example Sentences

But rather than teaching a diffusion model to generate images, the researchers teach it to generate a trajectory for a robot.

From Science Daily • Jun. 3, 2024

They represent each policy using a type of generative AI model known as a diffusion model.

From Science Daily • Jun. 3, 2024

Although AI has been in development for decades, the most popular of today's image generators use a technique called a diffusion model, which is relatively new on the AI scene.

From Scientific American • Sep. 22, 2023

In the recent study, the researchers used Stable Diffusion, a so-called diffusion model from London-based start-up Stability AI.

From Scientific American • Mar. 17, 2023

The diffusion model is trained on a series of images in which noise — imperfection — is gradually added to a photograph until it becomes a sea of random pixels.

From New York Times • Jan. 9, 2023