SoylentNews

Phoenix666 [soylentnews.org] writes:

Tiny flakes of graphene [ieee.org] may hold the key to building computer chips that can processes information similar to the way the human brain does [ieee.org]—only far faster—potentially leading to everything from better image recognition to control systems for hypersonic aircraft.

Researchers are developing so-called neuromorphic [ieee.org] chips consisting of networks of transistors that interact the way human neurons do, allowing them to process analog input, such as visual information, more quickly and accurately than traditional chips can.

One way of building such transistors is to construct them of lasers that rely on an encoding approach called “spiking.” Depending on the input, the laser will either provide a brief spike in its output of photons or not respond at all. Instead of using the on or off state of the transistor to represent the 1s and 0s of digital data, these neural transistors rely on the time intervals between spikes.

“We’re essentially using time as a way of encoding information,” says Bhavin Shastri [princeton.edu], a postdoctoral fellow in electrical engineering at Princeton University. Computation is based on the spatial and temporal positions of the pulses. “This is sort of the fundamental way neurons communicate with other neurons,” he says.

Original Submission