Skip to content
This repository has been archived by the owner on Mar 20, 2024. It is now read-only.

Latest commit

 

History

History
26 lines (15 loc) · 1.52 KB

README.md

File metadata and controls

26 lines (15 loc) · 1.52 KB

Preference Model

Learning What You Like From Your Likes

Video Intro

This app demonstrates how computational creativity can work together with you to create beautiful pictures.

In the beginning, you will see random pictures dreamed up by a neural network. This neural network is called a generator. The people on the images look natural, but they do not exist!

When you like some pictures, the AI will learn them. Then, the generator starts to create new images sharing the common traits of your likes.

You can vary the pictures less or more by changing the Variance.

Now you can select the best of the new pictures to create even more beautiful ones. Eventually, you can let the AI make the person of your dreams!

In mathematical terms, the AI finds the image maximizing the function of your personal preference using the cross-entropy method. The unusual thing about this is that this function is in your brain.

Online Demo

Check out the browser demo app. It works best on a PC with an NVIDIA graphic card. It may fail to run on some devices.

Credits

The generator is from the Progressive Growing of GANs for Improved Quality, Stability, and Variation by Tero Karras, Timo Aila, Samuli Laine, and Jaakko Lehtinen, licensed under the Creative Commons CC BY-NC 4.0 license.