MIT Has Built A Low-Power AI Chip For Mobile


Researchers at MIT have announced the creation of a new chip that is both energy efficient and able to perform powerful tasks related to artificial intelligence.

The chip is expected to enable "neural networks," those modeled on the brain, to be implanted into mobile devices.

According to MIT, neural networks tend to be implanted into the graphics processing unit within mobile devices. This is because a mobile GPU may have as many as 200 processing units or cores, making it a good option in the creation of a network.

This chip, however, is 10 times as efficient than other mobile GPUs, meaning that it could allow for mobile devices to run artificial intelligence algorithms without having to upload and download data from the Internet, where artificial intelligence data is currently processed.

"Right now, the networks are pretty complex and are mostly run on high-power GPUs. You can imagine that if you can bring that functionality to your cell phone or embedded devices, you could still operate even if you don't have a Wi-Fi connection," said Vivienne Sze, a professor at the MIT Department of Electrical Engineering and Computer Science, in a statement. "You might also want to process locally for privacy reasons. Processing it on your phone also avoids any transmission latency, so that you can react much faster for certain applications."

The new chip was dubbed by researchers as "Eyeriss" and could be useful for ushering in the Internet of Things, or the idea of everything being connected. With the ability to process things by themselves, connected devices could make important decisions on their own, uploading only final outcomes instead of all the data to the Internet.

It will certainly be interesting to see what the Internet of Things looks like in a few years, and how artificial intelligence interacts with the Internet. This chip and others like it could be a big boost to connected devices in general, especially for those who don't have fast Internet connections.

ⓒ 2018 All rights reserved. Do not reproduce without permission.
Real Time Analytics