Would you believe that artificial intelligence has gotten sophisticated enough to recreate an entire game just by analyzing footage of it? That’s exactly what NVIDIA researcher Seung-Wook Kim and his collaborators pulled off with GameGAN, a powerful new neural network model that’s capable of emulating the original PAC-MAN experience without an underlying engine. The feat was made possible by the magic of generative adversarial networks, which form data by leveraging competing systems.
“This is the first research to emulate a game engine using GAN-based neural networks,” said Kim, the lead author on the project. “We wanted to see whether the AI could learn the rules of an environment just by looking at the screenplay of an agent moving through the game. And it did.”
As NVIDIA points out, GameGAN has interesting implications for the future of game development. If someone wanted to generate hundreds of new levels, for instance, such AI models could save a ton of time, as they would need little human intervention.
“We were blown away when we saw the results, in disbelief that AI could recreate the iconic PAC-MAN experience without a game engine,” said Koichiro Tsutsumi from BANDAI NAMCO Research Inc., the research development company of the game’s publisher BANDAI NAMCO Entertainment Inc., which provided the PAC-MAN data to train GameGAN. “This research presents exciting possibilities to help game developers accelerate the creative process of developing new level layouts, characters and even games.”