Brain-On-A-Chip Gets Its Own Programming Language
August 12, 2013
When designing products or infrastructure, engineers often turn to nature. Architecture, medical devices, airplanes and other technologies are often inspired by the natural world. Since evolution has honed the design of natural systems over the course of billions of years, nature often provides highly-efficient solutions to problems.
In 2012, researchers working on the SyNAPSE project announced that they had successfully simulated 100 trillion synapses of a monkey brain on a powerful supercomputer, dubbed Sequoia. Instead of using hardware systems, researchers have now developed a neuromorphic chip, dubbed True North. Researchers also developed a specialized programming language to work with it.
For now, computers manipulate information by shifting it between a processor and memory. Classical computers like these are very efficient at crunching numbers. However, this approach doesn't work well for adaptive, creative intelligence. By taking a cue from biology, researchers believe computers can get better at activities like recognizing faces, images and color.
True North comprises a series of neurosynaptic cores that place communication, processing and memory close to each other so they can operate in parallel. During its early development, researchers had to code individual neurosynaptic cores for the system. However, the system now includes 150 specialized corelets for things like color identification, edge detection and sound perception.
"It doesn't make sense to take a programming language from the previous era and try to adapt it to a new architecture. It's like a square peg in a round hole. You have to rethink the very notion of what programming means," stated Dharmendra Modha, SyNAPSE manager.
About the Author
You May Also Like