MTL News

Neural networks everywhere

February 14, 2018

New chip reduces neural networks' power consumption by up to 95 percent, making them practical for battery-powered devices.

MIT-Neural-Network-Chip_0.jpg
MIT researchers have developed a special-purpose chip that increases the speed of neural-network computations by three to seven times over its predecessors, while reducing power consumption 93 to 96 percent. That could make it practical to run neural networks locally on smartphones or even to embed them in household appliances.

Image: Chelsea Turner/MIT

Read the full story.


This section of the Microsystems Technology Laboratories web site is intended to keep its community and outside readers informed about the activities of faculty and students who use its facilities. If you would like to post news or information on this page, call (617) 452-2545.

News & Events

MTL Annual Research Report 2017: View Online