Artificial intelligence (AI) researchers around the world will soon be able to use DeepMind’s “flagship” platform to develop innovative computer systems that can learn and think for themselves.
DeepMind, which was acquired by Google for £400 million in 2014, announced on Monday that it is open-sourcing its “Lab” from this week onwards so that others can try and make advances in the notoriously complex field of AI.
The company says that the DeepMind Lab, which it has been using internally for some time, is a 3D game-like platform tailored for agent-based AI research.
Founded in 2010, DeepMind has been developing AI agents that can master arcade games like “Space Invaders,” “Pac-Man,” and more recently the incredibly complex Chinese board game of Go.
Describing the Lab platform in a blog post, DeepMind cofounder Shane Legg and DeepMind employees Charles Beattie, Joel Leibo, Stig Petersen, wrote: “It is observed from a first-person viewpoint, through the eyes of the simulated agent. Scenes are rendered with rich science fiction-style visuals. The available actions allow agents to look around and move in 3D. The agent’s ‘body’ is a floating orb. It levitates and moves by activating thrusters opposite its desired direction of movement, and it has a camera that moves around the main sphere as a ball-in-socket joint tracking the rotational look actions.
“Example tasks include collecting fruit, navigating in mazes, traversing dangerous passages while avoiding falling off cliffs, bouncing through space using launch pads to move between platforms, playing laser tag, and quickly learning and remembering random procedurally generated environments.”
The DeepMind Lab aims to combine several different AI research areas into one environment. Researchers will be able to test their AI agent’s abilities on navigation, memory, and 3D vision, while determining how good they are at planning and strategy. “Each are considered frontier research questions in their own right,” DeepMind wrote in the blog post. “Putting them all together in one platform, as we have, represents a significant new challenge for the field.”
The Lab can be adapted and extended, with the possibility to create new “levels” that can be “customised with gameplay logic, item pickups, custom observations, level restarts, reward schemes, in-game messages and more.”
“We believe it has already had a significant impact on our thinking concerning numerous aspects of intelligence, both natural and artificial,” wrote the blog posts’ authors. “However, our efforts so far have only barely scratched the surface of what is possible in DeepMind Lab. There are opportunities for significant contributions still to be made in a number of mostly still untouched research domains now available through DeepMind Lab.”
DeepMind said the code for the DeepMind Lab platform is going to be published on Github this week.
The academic paper describing DeepMind Labs can be read here.