Exploring the power of generative adversarial networks
The Art of Forgery

© Lead Image © Aliaksandr-Marko, 123RF.com
Auction houses are selling AI-based artwork that looks like it came from the grand masters. The Internet is peppered with photos of people who don't exist, and the movie industry dreams of resurrecting dead stars. Enter the world of generative adversarial networks.
Machine learning models that could recognize objects in images – and even create entirely new images – were once no more than a pipe dream. Although the AI world discussed various strategies, a satisfactory solution proved illusive. Then in 2014, after an animated discussion in a Montreal bar, Ian Goodfellow came up with a bright idea.
At a fellow student's doctoral party, Goodfellow and his colleagues were discussing a project that involved mathematically determining everything that makes up a photograph. Their idea was to feed this information into a machine so that it could create its own images. At first, Goodfellow declared that it would never work. After all, there were too many parameters to consider, and it would be hard to include them all. But back home, the problem was still on Goodfellow's mind, and he actually found the solution that same night: Neural networks could teach a computer to create realistic photos.
His plan required two networks, the generator and the discriminator, interacting as counterparts. The best way to understand this idea is to consider an analogy. On one side is an art forger (generator). The art forger wants to, say, paint a picture in the style of Vincent van Gogh in order to sell it as an original to an auction house. On the other hand, an art detective and a real van Gogh connoisseur at the auction house try to identify forgeries. At first, the art expert is quite inexperienced, but the detective immediately recognizes that it is not a real van Gogh. Nevertheless, the counterfeiter does not even think of giving up. The forger keep practicing and trying to foist new and better paintings off on the detective. In each round, the painting looks more like an original by a famous painter, until the detective finally classifies it as genuine.
[...]
Buy this article as PDF
(incl. VAT)
Buy Linux Magazine
Subscribe to our Linux Newsletters
Find Linux and Open Source Jobs
Subscribe to our ADMIN Newsletters
Support Our Work
Linux Magazine content is made possible with support from readers like you. Please consider contributing when you’ve found an article to be beneficial.

News
-
Plasma 6.3 Now Available
Plasma desktop v6.3 has a couple of pretty nifty tricks up its sleeve.
-
LibreOffice 25.2 Has Arrived
If you've been hoping for a release that offers more UI customizations, you're in for a treat.
-
TuxCare Has a Big AlmaLinux 9 Announcement in Store
TuxCare announced it has successfully completed a Security Technical Implementation Guide for AlmaLinux OS 9.
-
First Release Candidate for Linux Kernel 6.14 Now Available
Linus Torvalds has officially released the first release candidate for kernel 6.14 and it includes over 500,000 lines of modified code, making for a small release.
-
System76 Refreshes Meerkat Mini PC
If you're looking for a small form factor PC powered by Linux, System76 has exactly what you need in the Meerkat mini PC.
-
Gnome 48 Alpha Ready for Testing
The latest Gnome desktop alpha is now available with plenty of new features and improvements.
-
Wine 10 Includes Plenty to Excite Users
With its latest release, Wine has the usual crop of bug fixes and improvements, along with some exciting new features.
-
Linux Kernel 6.13 Offers Improvements for AMD/Apple Users
The latest Linux kernel is now available, and it includes plenty of improvements, especially for those who use AMD or Apple-based systems.
-
Gnome 48 Debuts New Audio Player
To date, the audio player found within the Gnome desktop has been meh at best, but with the upcoming release that all changes.
-
Plasma 6.3 Ready for Public Beta Testing
Plasma 6.3 will ship with KDE Gear 24.12.1 and KDE Frameworks 6.10, along with some new and exciting features.