Today researchers at IBM’s Almaden research center unveiled the “biggest artificial brain ever,” which is described as being as large as a cat’s brain. The hardware is impressive, consisting of 147,456 CPUs, but what is even more compelling is how the team plans to “teach” the computer – by creating a virtual world in which it can explore and learn:
The simulation that Modha will unveil today is just a starting point. It lacks the neural patterning that develops as real brains mature. Neuroscientists believe that this complexity can only evolve through “embodied learning”—stumbling around in a physical body, in which every action has instant consequences that are experienced through senses such as touch and sight. As Anil Seth, a neuroscientist at the University of Sussex in Britain, puts it, “The brain wires itself.”
Seth demonstrated this principle while at the Neurosciences Institute in San Diego using a brain simulation called Darwin. He embodied Darwin’s 50,000 virtual neurons (about equal to the brain of a pond snail, or one-quarter of a fruit fly) in a wheeled robot. As Darwin wandered around, its virtual neurons rewired their connections to produce so-called hippocampal “place cells”—similar to neurons found in mammals—which helped it navigate. Scientists don’t know how to program these place cells, but with embodied learning the cells emerge on their own.
Paul Maglio, a cognitive scientist at Almaden, has similar plans for Modha’s cortical simulation. He’s building a virtual world for it to inhabit using software from the video shootout game “Unreal Tournament” and data from Mars. Besides topographic maps and aerial photos, Maglio plans to use rover-level imagery to create terrain with lifelike boulders and craters.
The video-game software provides a pallet of several dozen robotic bodies for Modha’s virtual cortex. Initially, it will use a simple wheeled robot to explore its world, driven by fundamental desires such as sustenance and survival. “It’s got to like some things and not like other things,” Maglio says. “Ultimately, it’s going to want not to roll off the edges of cliffs.”
Creating a virtual world for AIs has been suggested as a way to enable them to learn and interact with human avatars or other AIs without the danger of them making decisions that could affect the “real world.”
Of course there is no danger of this machine, even as powerful as it is, somehow gaining sentience and becoming a danger to others. In this case a virtual world is the only environment in which a machine of this size (which takes up a room the size of an acre) can learn from its actions in a way that roughly approximates the physical world. I expect this to be the norm until we’re able to come up with a way to shrink a machine with a comparable level of processing power down to a size that can practically exist as a robot – an event that is still decades away.
[...] week I wrote about IBM’s announcement that a research team had created the “largest artificial brain ever,” which they compared to [...]