A New Kind of Intelligence

Google DeepMind has successfully navigated a 3D maze without cheating—it didn’t have access to the digital world’s internal code. Instead, it walked around walls and into rooms by “sight,” as New Scientist reports.

Recently, DeepMind has successfully played the ancient game of “Go.” Now , it tackled a 3D maze much like the one in the 1993 Doom game. In the video below, the AI navigates through the maze simply by “sight” or by looking at the screen and deciding on what to do next, just as humans would play it.

How it works

The artificial intelligence relied on a technique called “reinforcement learning,” which rewards the system for taking actions that improves its score. This was combined with a deep neural network that analyses and learns patterns on the game screen. It was also able to look back into its memory and study past scenarios, a technique called “experience replay.”