Researchers at the U.S. Army Research Laboratory (ARL) used a low-cost, lightweight hardware system and implemented collaborative filtering on a low-power field-programmable gate array platform to achieve a 13.3-fold acceleration of training compared to a state-of-the-art, optimized multicore system.
ARL's Rajgopal Kannan says the method could be part of a suite of tools embedded on a next-generation combat vehicle, offering cognitive services and devices for soldiers in distributed coalition environments.
Kannan, who worked with researchers at the University of Southern California on this project, also is collaborating with them to develop other techniques to speed up artificial intelligence and machine learning algorithms via innovative designs on state-of-the-art (but inexpensive) hardware.
From U.S. Army Research Laboratory
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA
No entries found