English
Gamereactor
news

Great games by small devs: the magic of machine learning

Google Stadia's Erin Hoffman-John has been talking about machine learning can help power development in the future.

Subscribe to our newsletter here!

* Required field
HQ

Machine learning is a branch of Artificial Intelligence where the purpose is to create systems that automatically learn, both through patterns and algorithms, to self-correct and improve. For example, in some companies, machine learning is already used to detect future customers who will unsubscribe and try and prevent that using the data it has collected, all in an automated way.

In video games, there is still plenty of room for this discipline to develop on its own, and for this reason, Google is already beginning to study the tech. As Wccftech reports, Google is running Project Chimera, a group of engineers and programmers looking for applications to achieve this type of operation.

Google Stadia Creative Director, Erin Hoffman-John, has revealed to UK trade magazine MCVUK that machine learning could enable small teams to develop games as big as World of Warcraft.

"What if a team of 14 people could make a game the scale of World of Warcraft? That's an absurd goal, right? The thing about games like WoW is that they rely on a lot of heavy, repetitive content creation. The artists and the writers are doing a lot of essentially duplicate work, that's where a lot of the investment goes. If you look at the amount of money that is spent making a game like World Warcraft, it's like 70% content and 30% or less code."

The Google Chimera Project aims not only to facilitate the work of game developers but to find ways like this to be able to optimise a very complex creation around a small team.

Machine learning is already being tested in video games, although not yet in development. For example, DeepMind's AlphaStar AI is capable of winning 99.8% of Starcraft II players. Milestone is using it in ANNA, the AI that powers the machine-controlled riders in MotoGP 20. Ubisoft has also been experimenting with using it for bug-testing. All of this would be of great benefit to small studios that could aspire to the development of huge video games without the huge development costs.

Great games by small devs: the magic of machine learning


Loading next content