24/7 Customer Service (800) 927-7671

Google DeepMind has may already conquered the world of Go, but its upcoming accomplishment may be walking around in a game of Doom or GoldenEye 007.

by • February 4, 2016 • No Comments

In the maze, DeepMind is rewarded for finding apples and portals as it attempts to get a high score in only one minute. DeepMind moves around the labyrinth via a reward-based method — asynchronous reinforcement learning — and a neural network which recognizes patterns in the digital space. Yep, DeepMind in fact learns of its past experiences. But, the asynchronous method does not rely on examining previous run-throughs, a process which takes a ton of computing power. Instead, asynchronous reinforcement learning allows for the process to see multiple outcomes at once and select the many efficient path forward.

Google DeepMind utilized a much like reinforcement learning method in 2015 to play a handful of classic Atari games. The team has since improved and streamlined which program, enabling the AI to advance to the latest Doom-like maze.

Latest posts

by admin • March 5, 2017