AI recreates Pac-Man from scratch by just looking at the screenplay

To celebrate 40-years of Pac-Man, NVIDIA GameGAN, a new artificial intelligence model created a fully playable version of the all-time classic. The AI model was trained on 50,000 episodes of the game, it learned and created all this without the underlying game engine. This is amazing!

If we were to do something like this and we are good at drawing, we would have merely drawn a rough sketch of it…but creating the full working game is another level.

Generative adversarial networks or GANs, are made up of two competing neural networks, a generator and a discriminator. NVIDIA Research has the same as GameGAN which can mimic a computer game engine and create new content.

In a blog post about the project NVIDIA Research team member and lead author Seung-Wook Kim said, “This is the first research to emulate a game engine using GAN-based neural networks. 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.”

With GAN, AI model learns from game play episodes and recreates the screen elements such as maze shape, dots with Pac-Man himself. This also creates the game rules by just watching the element movements in the game.

This can be helpful for the developers to auto generate layouts for new game levels. This can also be really a game changer for simulating environments and recreate the virtual replica of the same. AI can suggest what can be changed in business environments to make processes much simpler or improve the overall output.

Possibilities are endless. Be ready to embrace the future today.

If you liked the story and want to see more of this kind of content, please follow us on twitter @CursorRun or FB @CursorRun

Default image
Rahul Sharma
Rahul is a technology enthusiast, solutions architect, trainer, and blogger, working on various Microsoft and open source solutions with more than 18 years of industry experience. He specially takes interest in designing enterprise applications, cloud integrations, IoT, and other architecture rich business solutions.

Leave a Reply

%d bloggers like this: