Welcome to the July 6, 2022, edition of ACM TechNews, providing timely information for IT professionals three times a week.
ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).
To view "Headlines At A Glance," hit the link labeled "Click here to view this online" found at the top of the page in the html version.
The online version now has a button at the top labeled "Show Headlines."
|
|
|
Acoustic Levitation Used to Build Complex Structures in Mid-Air
New Scientist
Matthew Sparkes
June 29, 2022
A system developed by researchers at Spain's Public University of Navarre incorporates a robotic arm that generates sound waves to levitate, rotate, and move objects in order to build complex structures in mid-air without touching the parts. The robotic arm, which employs 40-kilohertz sound waves, was used to build objects by manipulating droplets of glue or resin and small sticks. After assembly, a beam of ultraviolet light was applied to set the glue or cure the resin. The technique is similar to three-dimensional printing but can be used on liquids, powders, and hot or hazardous substances.
*May Require Paid Registration
|
Braverman Receives Refreshed Computer Science Medal
The New York Times
Kenneth Chang
July 5, 2022
The International Mathematical Union awarded four mathematicians under the age of 40 the latest quadrennial Fields Medals, while Princeton University's Mark Braverman, 38, received the Abacus Medal, a newer award modeled after the Fields Medal for young computer scientists. Braverman's work spans the fields of dynamics, information theory, economics, and game theory. One particular focus is communication complexity, which Braverman said "asks about the amount of resources you need to perform some computational task" when the computation demands sharing of information between different parties. His interest was in back-and-forth conversation, and he designed a protocol that minimized how much each party must reveal about what they knew.
*May Require Paid Registration
|
Robotic Arms Connected to Brain of Partially Paralyzed Man Allow Him to Feed Himself
Frontiers Science News
Peter Rejcek
June 28, 2022
Johns Hopkins University researchers led a team that developed a robotic system comprised of a brain-machine interface (BMI) and a pair of modular prosthetic limbs that allowed a partially paralyzed man to feed himself. The system uses muscle movement signals from the user's brain to control the robotic prosthetics. The man, unable to use his fingers for about three decades, gestured with his fists in response to computerized voice prompts, enabling the robotic limbs to slice a piece of cake with a knife and to align a cake-laden fork with his mouth. Said Johns Hopkins' Francesco Tenore, "This shared control approach is intended to leverage the intrinsic capabilities of the [BMI] and the robotic system, creating a 'best of both worlds' environment where the user can personalize the behavior of a smart prosthesis."
|
Building Explainability into Components of ML Models
MIT News
Adam Zewe
June 30, 2022
Researchers at the Massachusetts Institute of Technology (MIT) and cybersecurity startup Corelight have developed a taxonomy to help developers create components of machine learning (ML) models that incorporate explainability. The researchers defined properties that make features interpretable for five varieties of users, and that provide instructions for engineering features into formats that will be easier for laypersons to understand. Key to the taxonomy is the precept that there is no universal model for interpretability. The researchers define properties that can make components approximately explainable for different decision-makers, and outline which properties are likely most valuable to users. MIT's Alexandra Zytek said, "The taxonomy says, if you are making interpretable features, to what level are they interpretable? You may not need all levels, depending on the type of domain experts you are working with."
|
Algorithm Claims to Predict Crime in U.S. Cities Before It Happens
Bloomberg
Carrington York
June 30, 2022
University of Chicago social scientists have crafted a computer algorithm which, they say, can forecast crime in big U.S. cities a week ahead of time with 90% accuracy. Previous crime prediction models projected crime as originating from "hotspots" that expand to surrounding areas, which tends to overlook the complex social medium of cities, as well as the delicate relationship between crime and the impact of police enforcement, which can foster bias. The new algorithm segments cities into 1,000-square-foot tiles, detecting patterns over time in these areas. It uses hundreds of thousands of sociological patterns to ascertain the risk of crime at a particular time and space.
*May Require Paid Registration
|
Researchers Invent Bluetooth That Keeps Working Even If Power Runs Out
TU Delft (Netherlands)
June 22, 2022
Jasper de Winkel and Przemyslaw Pawelczak at the Netherlands' Delft University of Technology have enabled energy-efficient two-way Internet of Things (IoT) communication via intermittently-powered Bluetooth. De Winkel said the FreeBie implementation puts the Bluetooth chip "in sleep mode whenever the IoT device is idle. By saving the exact state, however, we can switch the chip off completely, reducing energy consumption by an order of magnitude." The researchers used commercially available hardware to implement FreeBie, and deployed two versions: a solar-powered battery-free smartwatch that synchronizes with a smartphone’s time display and email notifications, and Bluetooth-facilitated wireless firmware updates. Said de Winkel, "We hope that it will open the door for other wireless protocols: intermittently-powered Wi-Fi, 5G, 6G, you name it."
|
Toe-Tapping Test for Parkinson's Patients
IEEE Spectrum
Michelle Hampson
June 30, 2022
Texas A&M University's Ya Wang and colleagues devised a smart insole that can evaluate a Parkinson's disease patient's risk of falling by tracking their foot movements as they tap their toes. The researchers had 10 Parkinson's patients and eight healthy controls perform a toe-tapping exercise while wearing the insole, which features accelerometers, a microcontroller, and a Bluetooth Low Energy module. The device recorded and transferred data on foot motion to a computer in real time. The researchers found the exercises strongly correlate with falling risk based on the Unified Parkinson's Disease Rating Scale, although with slightly less accuracy than conventional walking tests.
*May Require Free Registration
|
NIST Identifies Quantum-Resistant Encryption Algorithms
Nextgov
Alexandra Kelley
July 5, 2022
Officials from the U.S. National Institute of Standards and Technology (NIST) announced the first four quantum-resistant encryption algorithms, dubbed Crystals-Kyber, Crystals-Dilithium, Falcon, and SPHINCS+. The announcement marks the start of the final phase of NIST's research into the development of a post-quantum cryptographic standard to shield digital information against quantum hacking. U.S. Commerce Secretary Gina Raimondo said NIST's achievement means "we are able to take the necessary steps to secure electronic information so U.S. businesses can continue innovating while maintaining the trust and confidence of their customers." NIST director Laurie E. Locascio said this initial slate of quantum-resistant algorithms "will lead to a standard and significantly increase the security of our digital information."
|
Eavesdropping on Whales in the High Arctic
Norwegian SciTech News
Nancy Bazilchuk
July 5, 2022
Norwegian researchers used existing underwater fiber-optic cables to listen to whales in the Arctic via the Distributed Acoustic Sensing (DAS) system. "With this system, which is what we can basically call a hydrophone array, we have the chance to cover a much bigger area for monitoring," said former Norwegian University of Science and Technology (NTNU) researcher Léa Bouffaut. "And because we receive the sound at multiple angles, we can even say where the animal was—the position of the animal." The researchers listened to underwater sounds in the Isfjorden region of the Svalbard archipelago between Norway and the North Pole, amassing a total of 7 terabytes of data each day for analysis. NTNU's Hannah Joy Kriesell suggested a DAS setup enabling real-time analysis could reduce the risk of ship-whale collisions.
|
Blockchain Can Secure, Store Genomes
YaleNews
Bill Hathaway
June 29, 2022
The SAMchain technology developed by Yale University scientists leverages blockchain to give users control over their own genomic data. SAMchain guarantees the security of individual genomic information, shielding it against change by others and the occasional corruption of cloud-stored DNA data. The researchers circumvented the problem of storing vast datasets derived from genome sequencing by comparing an individual's DNA against a standard reference genome, then storing only the differences in linked blocks of the blockchain. The blocks are subsequently indexed to enable rapid inquiry, and those differences can be connected to conditions with known genetic risk factors. "We think this will actually make genomic research easier," said former Yale researcher Gamze Gürsoy.
|
Using Google Search Data to Forecast COVID Hospitalizations
Voice of America News
Elise Cutts
June 29, 2022
Georgia Institute of Technology (Georgia Tech) scientists used search data from Google Trends to build a predictive computer model of U.S. COVID-19 hospitalizations. Georgia Tech's Shihao Yang said the model’s perspective is that of a data source that is independent of conventional metrics. The model gauges Google search volumes for 256 COVID-19-specific terms, including "loss of taste," "COVID-19 vaccine," and "cough," plus core statistics such as case counts and vaccination rates. Also incorporated into the model are terms representing the delay between one day's data and future forecast hospitalizations, and how closely different states correspond. The model retrains itself every week using the past 56 days' worth of data, which keeps it from being overloaded with older data that does not reflect current virus trends.
|
'Fake' Data Helps Robots Learn the Ropes Faster
University of Michigan News
June 29, 2022
University of Michigan (U-M) researchers expanded training datasets for teaching robots to work with soft objects like ropes and fabrics, to expedite the learning process. U-M's Dmitry Berenson and Peter Mitrano augmented an optimization algorithm to enable a computer to make human-level generalizations, forecasting how dynamics observed in one case might recur in others and generating variants of the first experiment's result that serve the robot in the same way. In simulations, the expanded dataset improved the success rate of a robot looping a rope around an engine block by 48%, which increased to 70% after training. An experiment using a physical robot to perform the same task almost doubled its success rate over the course of 30 tries.
|
A 'Sunflower' Eye in the Sky
Dartmouth News
Harini Barath
June 30, 2022
Dartmouth College researchers have created Sunflower, an aerial sensor system that uses lasers which enable aerial drones to detect and locate aquatic robots. A flying drone (queen) scans the water with a laser beam; the aquatic robots (workers) have reflectors that bounce the laser light back to the drone, which captures data including their depth and the angle at which the light struck them. Dartmouth's Charles Carver said this allows the queen to compute each worker's position. Tests in a swimming pool with simulated waves and wind showed the drone could pinpoint the robots with centimeter-level accuracy.
|
|
