Researchers at IBM Have Developed a Chip Inspired by the Human Brain
In a project that started back in 2008, researchers at IBM have developed a computer chip that works very similarly to the way that neurons work in our brains. As a part of DARPA’s Systems of Neuromorphic Adaptive Plastic Scalable Electronics (SyNAPSE) project, IBM has released their progress with the TrueNorth system.
Buy an “Intelligence is sexy” t-shirt!
The TrueNorth system is actually a network comprised of 48 chips that each has 1 million artificial nerve cells. The array of chips is described as “neuromorphic,” which means they are designed to work much like the human brain. IBM was quick to point out that IBM researchers note that they “have not built the brain, or any brain” but have built “a computer that is inspired by the brain.” Traditional computers work like the left side of our brains which is the “calculator-like” side. The TrueNorth system is designed to work more like the right side of our brains which is slower, but more analytical. This makes the TrueNorth system great for things like facial recognition or instant translation.
The problem with similar systems in the past is that they are expensive, and require a ton of power to run. According to Wired, TrueNorth’s 5.4-billion transistor chip uses 70 milliwatts of power in comparison to a standard Intel processor with 1.4 billion transistors that uses about 35 to 140 watts. According to Brian Van Essen, a computer scientist at the Lawrence Livermore National Laboratory, “What does a neuro-synaptic architecture give us? It lets us do things like image classification at a very, very low power consumption. It lets us tackle new problems in new environments.”
Read: The Spectacular Science Behind Sonogenetics: Controlling the Brain Through Sound
Although it will be several years before TrueNorth’s technology is available to the public, IBM says says these chips will solve “a wide class of problems from vision, audition, and multi-sensory fusion, and has the potential to revolutionize the computer industry by integrating brain-like capability into devices where computation is constrained by power and speed.”