AIs Battle to the Death in Ms. Pac-Man Competition

This article is over 13 years old and may contain outdated information

Recommended Videos

The results of 2011’s Ms Pac-Man vs Ghost Team Competition were announced today by Philipp Rohlfshagen and Simon Lucas at the Congress on Evolutionary Computation in New Orleans. Now, this wasn’t some namby-pamby person-on-person video game break organized for the conference; this was an all-out death match in which entrants wrote artificial intelligence programs to control either Ms. Pac-Man or the Ghost Team.

In the competition, each AI controlled competitor moved through a gauntlet of matchups until the dust settled and one winner stood out amongst the carnage. The roles were fairly straightforward: Ms. Pac-Man aims to get the highest score, while the Ghost Team AIs are designed to suppress the plucky puck-shaped heroine. The competition was brutal, since the highest Ms. Pac-Man score in the competition was 69, 204 — paltry compared to the 900,000 achieved by human players on conventional machines. But keep in mind, conventional machines are designed to be fun to play, not cold-hearted and cunning.

Of course, it’s not all just for fun. While a Pac-Man playing AI is of limited application, the process of developing and testing one could have implications for other projects. From the New Scientist:

“Games are usually seen as a valuable test-bed for new technologies in computational intelligence as they are well defined yet very challenging,” explains Rohlfshagen. He says the multi-agent algorithms behind the ghost controllers could be used for transport or military applications, or even modelling biological predator-prey dynamics.

Perhaps, one day, a swarm of futuristic military drones will be able to trace its lineage to the humble and ever vigilant Ghost Team.

(via New Scientist)


The Mary Sue is supported by our audience. When you purchase through links on our site, we may earn a small affiliate commission. Learn more about our Affiliate Policy
Author
Image of Max Eddy