Technology continues to open up new horizons with groundbreaking developments. One of those recent developments comes from computer maker International Business Machines Corp. (NYSE:IBM), which just unveiled a new brain-like chip made from 4,000 processor cores.
A new innovation
Scientists have long searched for ways to create innovative technologies that mimic the human brain. It was only in the more recent years scientists made strident efforts in creating neuro-like computer chips. These chips were designed to process information differently than traditional computer chips.
This year has been a record-breaking year for initial public offerings with companies going public via SPAC mergers, direct listings and standard IPOS. At Techlive this week, Jack Cassel of Nasdaq and A.J. Murphy of Standard Industries joined Willem Marx of The Wall Street Journal and Barron's Group to talk about companies and trends in Read More
IBM published a report in academic journal Science which goes further into detail about TrueNorth about the custom brain-like chip dubbed TrueNorth. This chip was actually built upon a simpler experimental system launched three years ago. The TrueNorth system features 4,096 processor cores. This helps it mimic 256 million synapses and one million human neurons, both of which are key building blocks which comprise the human brain. The computer maker termed these as “spiking neurons”, meaning the chip can encode data as pulses of patterns, a similar pattern researchers believe the brain stores knowledge.
How IBM’s new chips will change technology
Carver Mead (professor emeritus of engineering and applied science at California Institute of Technology) claims this is a neat experiment in architecture. He added traditional processors such as CPUs and GPUs are not very good at encoding data in a brain-like method. This where IBM’s new chips could prove useful by representing information by timing nerve pulses. This is not something digital computers were able to do in the past.
The tech company already tested the chip’s ability to use artificial intelligence such as recognizing images. The chip’s neurons and synapses handle those tasks with the same amount of speed but less power than most off-the-shelf chips. Researchers took it to the next level with DARPA’s NeoVision2 Tower dataset which included images taken from video recording taken atop the Hoover Tower at Stanford University. The TrueNorth chip recognized people, vehicles, and buses all with an accuracy level of 80 percent. In another test, researchers fed streaming video from TrueNorth at 30 frames of second which burned only 63 mW of power processing data in real-time.
Although these new chips have a lot of potential, critics claim this technology must pass many more tests before it can be utilized in mobile technology, cameras, or even robots.