A single silicon beam (red), along with its drive (yellow) and readout (green and blue) electrodes, implements a MEMS capable of nontrivial computations.
CREDIT: Guillaume Dion
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.
Learn more: New Reservoir Computer Marks First-Ever Microelectromechanical Neural Network Application
The Latest on: Reservoir computer
via Google News
The Latest on: Reservoir computer
- EDITORIAL: A River Runs Through It, Part Fiveon August 4, 2022 at 6:10 am
It doesn’t, however, get returned directly to the San Juan River following treatment. The Vista plant outflow spills into the Borns Draw, which then flows into Martienz Creek (when the outflow is ...
- Thirteen meters below its normal level: In the Alps, the Serre-Ponçon Lake is trying to adapt to droughton July 31, 2022 at 4:00 pm
In the Alps, th Lake is trying to adapt to drought By Audrey Garric Published on August 1, 2022 at 02h00, updated at 09h01 o ...
- A Colorado River Glossary: Jargon Explainedon July 28, 2022 at 9:20 am
Discussions about the Colorado River can resemble a doctor’s visit: filled with technical jargon, unfamiliar acronyms, and the anxiety that comes from incomplete understanding. This glossary is an ...
- State auditor investigates California's disappearing water—as a new culprit emergeson July 27, 2022 at 7:18 am
After years of drought and extreme heat, the environment is thirsty and sponging up moisture—creating a new adversary in California’s water wars.
- California reservoirs continue to struggleon July 25, 2022 at 1:06 am
It’s over 80% for many communities in the desert regions of Arizona and over 60% for Southern California residents. In addition for drinking water and irrigation to farmlands, Lake Mead also serves as ...
- Scientists use new techniques to identify lakes and reservoirs worldwideon July 19, 2022 at 11:36 am
The study highlighting the Reservoir and Lake Surface Area Timeseries ... in the University of Minnesota Twin Cities Department of Computer Science and Engineering. "This new dataset greatly ...
- Data scientists use new techniques to identify lakes and reservoirs around the worldon July 19, 2022 at 9:59 am
A paper highlighting the Reservoir and Lake Surface Area Timeseries ... in the University of Minnesota Twin Cities Department of Computer Science and Engineering. "This new dataset greatly ...
- Computer Modelling Group Announces Voting Results Of Election Of Directorson July 7, 2022 at 6:11 pm
Computer Modelling Group Ltd. is a computer ... CMG, recognized by oil and gas companies worldwide as a leading developer of reservoir modelling software, has sales and technical support services ...
- Computer Modelling Group Ltd. (TSE:CMG) Director Kenneth Michael Dedeluk Sells 200,000 Shares of Stockon June 14, 2022 at 1:32 am
Computer Modelling Group Ltd., a computer software technology company, develops and licenses reservoir simulation software in Canada and internationally. The company offers CMOST-AI, an ...
via Bing News