Achieving a perfect score in the 1980s video game Ms. Pac-Man is something to brag about, which is probably what Microsoft's artificial intelligence is doing right now, assuming it actually knows how to brag.
A team of engineers from Microsoft's deep learning subsidiary Maluuba wrote software for the Atari 2600 version of Ms. Pac-Man, dividing up the notoriously challenging game into separate parts for their algorithms to solve individually. Together, the AI components achieved the maximum possible score of 999,990, Microsoft announced on Wednesday.
Maluuba's AI architecture mimics the organizational structure of a company: there's a "top AI agent" who receives suggestions from all the other algorithms. It ranks those suggestions based not only on which direction they recommend Ms. Pac-Man to move, but also how confident the algorithms feel in making their decisions.
"For example, if 100 agents wanted to go right because that was the best path to their pellet, but three wanted to go left because there was a deadly ghost to the right, it would give more weight to the ones who had noticed the ghost and go left," Microsoft explained in a blog post.
More From PCmag
Ms. Pac-Man is a more complex game than the original Pac-Man, and no human has ever achieved its maximum possible score without cheating, according to The Verge. Of course, cutting up the game into pieces and feeding it to multiple AI algorithms might itself be considered cheating, so the monumentality of Microsoft's achievement comes down to whether or not you accept the researchers' methods.
Still, it's clear that beating Ms. Pac-Man is the latest in a string of game achievements for the AI divisions of Silicon Valley giants. Google's AlphaGo algorithm beat the human world champion of the ancient Chinese game Go in a match last month, the second time it has stolen the Go crown from a human opponent.