Nvidia AI Recreates Pac-Man Without Human Input

Nvidia AI Recreates Pac-Man Without Human Input

Today is Pac-Man’s 40th birthday and many companies are celebrating the icons special day in a variety of ways. Nvidia and Bandai Namco partnered up to recreate the classic game, but this time without any actual human input.

Utilizing Nvidia’s GameGAN, the company has successfully trained an AI model to recreate and play Pac-Man without the use of a game engine. The AI is actually recreating the game from scratch manually on screen while tracking the movement of the yellow hero, the pellets he eats, the ghosts he destroys, and the score it racks up.

The AI was trained using millions of frames from the original Pac-Man, which were supplied by Bandai Namco, along with recorded player inputs. After around 50,000 games the AI was able to recreate static elements from a simple Pac-Man game, while simultaneously understanding how to operate within its rules, without a single line of code entered by developers.

Pac Man

This is an awesome homage to one of the gaming industry’s pillars. This is also an excellent view into the future of level design and how AI will influence it. Nvidia and various researchers in the field believe this architecture could be used to add new content to a level that fits with what already exists. It also has benefits in both virtual and real world applications to teach AI the simplicities of human behaviors, to eventually recreate them in a game or be used in self-driving cars.

For now though, this is simply a cool look into the innovations occurring in the AI field and how Nvidia is playing a role in it. We’ll be sure to keep an eye out for any developments in this technology so be sure to keep checking back with us here at GigamaxGames.com!

Recent Post


EwinRacing Gaming Chairs


Use code Gigamax at checkout to get 20% off EwinRacing gaming chairs



Archive

  • July 2021
  • June 2021
  • May 2021
  • March 2021
  • February 2021
  • January 2021
  • December 2020
  • November 2020
  • October 2020
  • September 2020
  • August 2020
  • July 2020
  • June 2020
  • May 2020
  • April 2020
  • March 2020
  • February 2020
  • January 2020
  • December 2019
  • November 2019
  • October 2019
  • September 2019
  • August 2019
  • July 2019
  • June 2019
  • May 2019
  • April 2019
  • March 2019
  • February 2019
  • January 2019
  • December 2018
  • November 2018
  • October 2018
  • September 2018
  • August 2018
  • July 2018
  • June 2018
  • May 2018
  • April 2018
  • March 2018
  • February 2018
  • January 2018
  • December 2017
  • November 2017
  • October 2017
  • September 2017
  • August 2017
  • July 2017
  • June 2017
  • May 2017
  • April 2017
  • March 2017
  • February 2017
  • January 2017
  • December 2016
  • November 2016
  • October 2016
  • September 2016
  • August 2016
  • July 2016