Researchers used oscillations from a microscopic beam of silicon to enable the nonlinear dynamics that allow neural networks to complete tasks ranging from processing image patterns to recognizing words.
As artificial intelligence has become increasingly sophisticated, it has inspired renewed efforts to develop computers whose physical architecture mimics the human brain. One approach, called reservoir computing, allows hardware devices to achieve the higher-dimension calculations required by emerging artificial intelligence. One new device highlights the potential of extremely small mechanical systems to achieve these calculations.
A group of researchers at the Université de Sherbrooke in Québec, Canada, reports the construction of the first reservoir computing device built with a microelectromechanical system (MEMS). Published in the Journal of Applied Physics, from AIP Publishing, the neural network exploits the nonlinear dynamics of a microscale silicon beam to perform its calculations. The group’s work looks to create devices that can act simultaneously as a sensor and a computer using a fraction of the energy a normal computer would use.
The article appears in a special topic section of the journal devoted to “New Physics and Materials for Neuromorphic Computation,” which highlights new developments in physical and materials science research that hold promise for developing the very large-scale, integrated “neuromorphic” systems of tomorrow that will carry computation beyond the limitations of current semiconductors today.
“These kinds of calculations are normally only done in software, and computers can be inefficient,” said Guillaume Dion, an author on the paper. “Many of the sensors today are built with MEMS, so devices like ours would be ideal technology to blur the boundary between sensors and computers.”
The device relies on the nonlinear dynamics of how the silicon beam, at widths 20 times thinner than a human hair, oscillates in space. The results from this oscillation are used to construct a virtual neural network that projects the input signal into the higher dimensional space required for neural network computing.
In demonstrations, the system was able to switch between different common benchmark tasks for neural networks with relative ease, Dion said, including classifying spoken sounds and processing binary patterns with accuracies of 78.2 percent and 99.9 percent respectively.
“This tiny beam of silicon can do very different tasks,” said Julien Sylvestre, another author on the paper. “It’s surprisingly easy to adjust it to make it perform well at recognizing words.”
Sylvestre said he and his colleagues are looking to explore increasingly complicated computations using the silicon beam device, with the hopes of developing small and energy-efficient sensors and robot controllers.
The Latest on: Reservoir computer
via Google News
The Latest on: Reservoir computer
- Climate change presents new challenges for the drinking water supplyon November 23, 2020 at 10:39 am
The reservoir's operators draw the raw water from the lowermost strata for good reason, as the water there is cold and contains only low levels of suspended substances, dissolved metals, algae ...
- Joe's Fishing Hole: Cold temps send fish to shallow wateron November 20, 2020 at 8:34 am
With the colder weather comes colder water. Makes sense, doesn’t it? Some area waters are seeing some ice, especially at higher elevations. The beaver ponds in Lamoille Canyon are iced ...
- Mapping a wildfire's path is getting easier, thanks to computerson November 20, 2020 at 4:12 am
Fire-predicting software can project how a fire could spread, even while it's still burning. The wildfire burned through pine trees only a few miles away from Martin City, Montana, just outside of ...
- Computer Modelling Group has upside, Industrial Alliance sayson November 18, 2020 at 5:30 pm
Ken Dedeluk, President and CEO of Computer Modelling Group It’s been a rough year ... Calgary-headquartered CMG is a software and consulting company with dynamic reservoir modelling and simulation ...
- Joe's Fishing Hole: Learn fly-tying this winteron November 13, 2020 at 7:59 am
For many anglers, winter is a time to take a break from fishing and look to other interests that can be done inside. For fly fishermen and fisherwomen, winter is ...
- Computer Modelling Group Declares Quarterly Dividendon November 12, 2020 at 1:00 pm
The Board of Directors of . ("CMG" or the “Company”) announces a dividend of $0.05 per Common Share on CMG’s Common Shares. The dividend will be paid on December 15, 2020 to shareholders of record at ...
- Father of journeyson November 6, 2020 at 5:32 pm
I was carrying an introduction from Gopalkrishna Gandhi, who had described him to me as a ‘great Gandhi-reservoir’. For the next twenty-five years I was witness to the truth of that appellation, some ...
- CMDXFon November 6, 2020 at 4:00 pm
Computer Modelling Group Ltd., a computer software technology company, develops and licenses reservoir simulation software in Canada. The company offers CMOST AI, an intelligent optimization and ...
- Larkooler SkyWater 330 Liquid Cooling System CPU Cooler Reviewon November 2, 2020 at 5:01 pm
The connected reservoir will house 250mL of coolant ... With a few years of abusing computer parts, he decided to take his chances and try to get a review job. As an avid overclocker, Chad ...
via Bing News