Home → News → Georgia Tech Researchers Train Computer to Create... → Full Text

Georgia Tech Researchers Train Computer to Create Games By Watching Youtube

By Georgia Tech News Center

July 1, 2015

[article image]

Georgia Institute of Technology (Georgia Tech) researchers have developed a computing system that views video game play on streaming services, analyzes the footage, and then creates original new sections of a game.

The researchers tested the system with the original Super Mario Brothers game. The system focuses on the gaming terrain and the positioning between elements on-screen, and determines the required relationship or level design rule.

"An initial evaluation of our approach indicates an ability to produce level sections that are both playable and close to the original without hand coding any design criteria," says Georgia Tech researcher Matthew Guzdial.

The system relies on studying players in action to see where they spend most of their time in the game. The algorithms identify high-interaction areas, and the automatic-level designer targets these areas to gain design information, enabling the system to build a new level section, element by element.

"Our system creates a model or template, and it's able to produce level sections that have never been seen before, do not appear random, and can be traversed by the player," says Georgia Tech professor Mark Riedl.

The system created 151 distinct level sections from 17 samples of the original game, controlling for overall playability and style variables.

From Georgia Tech News Center
View Full Article


Abstracts Copyright © 2015 Information Inc., Bethesda, Maryland, USA


No entries found