The Physics Principle That Inspired Modern AI Art

Sohl-Dickstein used the principles of diffusion to develop an algorithm for generative modeling. The idea is simple: The algorithm first turns complex images in the training data set into simple noise—akin to going from a blob of ink to diffuse light blue water—and then teaches the system how to reverse the process, turning noise into images.

Here’s how it works: First, the algorithm takes an image from the training set. As before, let’s say that each of the million pixels has some value, and we can plot the image as a dot in million-dimensional space. The algorithm adds some noise to each pixel at every time step, equivalent to the

→ Continue reading at Wired - Science

More from author

Related posts

Advertisment

Latest posts

The Tantalizing Mystery of the Solar System’s Hidden Oceans

The original version of this story appeared in Quanta Magazine.For most of humankind’s existence, Earth was the only known ocean-draped world, seemingly unlike any...

The Role of Scholarships and Grants in Financing Education

Applying for scholarships and grants helps students pay for college. In some situations, these awards allow people to obtain a degree they wouldn't be...

EV, hybrid and gas-powered: Some interesting cars coming in 2024 | CNN Business

CNN  —  Next year will see the introduction of some new, genuinely affordable electric vehicles as well as...