PAC-MAN is recreated from scratch by NVIDIA Artificial Intelligence

Developed by NVIDIA, GameGAN manages to recreate the classic Pac-Man without using a game engine

When an old video game has its birthday, an increasingly common type of homage made by programmers is to teach an AI how to play it. But for the celebration of 40 years of PAC MAN, a NVIDIA took this kind of homage to another level, and not just taught a IA to play the game, but rather developed one that was able to create its own version of the game without the aid of any game engine.

Released in May 1980, the first PAC MAN was a huge success of arcades and one of the main responsible for the popularization of video games. Throughout its 40 years of history, PAC MAN it went from being just a game to one of the most famous and successful franchises in video game history, selling some 48 million copies distributed across multiple platforms and earning about US$14.10 billion in revenue.

And it’s to celebrate the game’s 40 years of success that the NVIDIA developed artificial intelligence GameGAN, a model of IA that manages to develop a fully playable version of PAC MAN just watching videos of people playing the game.

What is GameGAN from PAC-MAN

THE GameGAN is IA created by NVIDIA which was developed from adversarial generative networks (generative adversarial networks, in English) . Also known by the acronym GAN, this is a machine learning method that uses two neural networks that compete with each other, being a generator and a discriminator, to analyze a content and create a new one that is practically identical to the original.

Recommended
Zoom: Tips and tricks to improve your video calls

Created in 2014 by Ian Goodfellow (an AI researcher who currently works for the Apple), at AIs of the GAN type were developed to show that computer programs can learn to copy something even if they are not aware of the gears behind its operation, more or less like the human brain, which can make us imitate something even without understanding the whole basis behind that—like copying a pop artist’s dance choreography even if we don’t have any knowledge of music theory or even take dance lessons.

Thus, a team of researchers from the NVIDIA led by Seung-Wook Kim made the AI ​​“watch” the millions of possible frames (that is, every move the player could make in the game) of the phases of the PAC MAN original, in addition to the commands used by another IA who had been trained to play the game, and then they checked that the GameGAN could create a game equal to the original just from that experience.

The result was better than expected: a GameGAN not only managed to exactly recreate the entire game environment (the marble-filled maze with the PAC MAN, four bigger balls in the corners and four ghosts), but also understand practically all its logic, from the simplest ones (like the fact that the character doesn’t go through walls and needs to eat all the marbles on the map to pass the level) to the more complex (like the fact that the ghosts must always chase the player, but if the protagonist eats one of the bigger balls, these ghosts must all turn blue and start to run away for a certain period of time).

The future of the gaming industry

Although it wasn’t the first IA of the GAN type to be developed, the GameGAN gives NVIDIA was the first created for specific use in video games, and the result obtained can be used as the basis for an industry-wide change.

Recommended
League of Legends: Riot denies that Seraphine was inspired by a real young man

This is because, if new experiments with the GameGAN can achieve results as promising as those obtained in the game PAC MAN when using the IA in more complex games, it can become a powerful tool for game developers, helping them to create new scenarios and characters much faster, reducing the time needed to develop a video game.

And the same technology can be used for other functions as well. According to Sanja Fidler, director of the research laboratory at NVIDIA in Toronto, the GameGAN can be the first step in developing a IA that could learn how to drive a car just watching videos of people doing this activity and how they behave in different traffic situations, which could be a new approach to the development of autonomous vehicles.

For now, the NVIDIA is just releasing the results obtained by the survey, but has already promised that by the end of the year it will release a fully playable version of PAC MAN developed by GameGAN at the IA Playground, the platform where company researchers make available for free some programs developed in their laboratories that use artificial intelligence.

Source: NVIDIA

Related Articles

Back to top button

Adblock Detected

Please consider supporting us by disabling your ad blocker. Thanks.