Welcome to the March 14, 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."
|
|
|
Citizen Seismologists Track Haiti's Earthquakes
The New York Times
Kenneth Chang
March 10, 2022
Citizen scientists recorded a magnitude-7.2 earthquake in Haiti last August using inexpensive seismometers, providing critical information in real time, said Eric Calais at France's École Normale Supérieure. Susan E. Hough with the U.S. Geological Survey said the lack of a functional power grid and reliable Internet in Haiti made operating a conventional cutting-edge seismic network very difficult. The researchers utilized a device from Panama-based company Raspberry Shake that combines a Raspberry Pi computer with an inexpensive gadget for measuring tiny ground movements. Calais and colleagues recruited volunteers in Haiti to use the mini-seismometers, which Calais said made scientifically valuable measurements, including small aftershocks. Six Raspberry Shakes were deployed in all, measuring 1,000-plus aftershocks in the weeks following the quake.
*May Require Paid Registration
|
Robots with Realistic Pain Expressions Can Reduce Examination Error, Bias
Imperial College London (U.K.)
Alex Dadswell
March 11, 2022
Scientists at the U.K.'s Imperial College London have developed a technique for generating more realistic facial expressions of pain on robotic patients, which could help identify and correct signs of bias among medical students. Undergraduates performed physical examinations on the abdomen of a robotic patient, as data about the force applied to the belly triggered changes in regions of the robotic "MorphFace" to replicate pain-related expressions. The method showed the order in which these facial activation units (AUs) must initiate to produce the most accurate pain expression, as well as the most suitable speed and magnitude of activation. The researchers determined the most realistic expressions occurred when AUs around the eyes, then those around the mouth, were activated. Participants also noted perception of the extent of the robot patient's pain depended on gender and ethnic differences between participant and patient, which impacted the force applied during examination.
|
AR Models Bring Ice Age Animals to Virtual Life
Scientific American
Sam Jones
March 10, 2022
Researchers at the Natural History Museum of Los Angeles County, the La Brea Tar Pits and Museum, and the University of Southern California (USC) developed three-dimensional (3D) animated models of 13 Ice Age animals whose fossils were found in the La Brea Tar Pits. In conjunction with a video game development company, the models were adapted for use in augmented reality (AR)-enhanced museum exhibits and smartphone-based AR experiences on social platforms. The animals were recreated using low-polygon graphics, which allow the researchers to use the most up-to-date paleontological findings about the animals' appearances and behaviors, and avoid inaccuracies associated with guessing at things the fossils do not reveal.
|
CS Professor’s 'Hands-On' Approach to Smartphone Security
LSU College of Engineering
March 9, 2022
Louisiana State University computer science professor Chen Wang is working with Ph.D. student Long Huan on a smartphone security system that only displays sensitive content when it verifies the correct user is holding the device. The gripping-hand verification method employs an artificial intelligence (AI)-based algorithm that processes the notification tone recorded by the phone's mic, extracts biometric features to match with the user's recorded hand grip, and displays the notification preview if the verification is successful. Wang explained, "Because people have different hand sizes, finger lengths, holding strengths, and hand shapes, the impacts on sounds are different and can be learned and distinguished by AI. Along this way, we develop a system to use the notification tones to verify the gripping hand for notification privacy protection."
|
Walgreens Replaced Some Fridge Doors with Screens; Some Shoppers Absolutely Hate It
CNN Business
Nathaniel Meyersohn
March 12, 2022
Some customers are furious with Walgreens and other retailers’ replacement of clear fridge and freezer doors with iPad-like screens displaying items within. The screens use motion sensors and cameras to display the contents of the refrigerated shelves along with prices, and even ads. Shoppers have expressed annoyance and confusion, especially with the ads. The University of Georgia's Julio Sevilla said digital screens can add uncertainty and physical obstruction, and he doubts consumers seek novelty when shopping for groceries. Walgreens has deployed the digital screens at thousands of its U.S. outlets, while Kroger, CVS, GetGo convenience stores, and Chevron gas stations are launching their own tests of the screens.
|
SMASH Lab Uses Wearables to Train Privacy-Preserving Sensors
Carnegie Mellon University School of Computer Science
Aaron Aupperlee
March 10, 2022
Researchers at Carnegie Mellon University's Smart Sensing for Humans (SMASH) Lab are using data collected from smartwatch inertial measurement unit (IMU) sensors to train a millimeter wave doppler to identify human movements and behaviors. The IMU2Doppler ambient sensor could be used in smart houses to track daily activities like eating, drinking, brushing teeth, and folding clothes, while maintaining users' privacy. Eventually, doppler sensors could help medical professionals monitor patients at home to detect early signs of illness, or other issues. The researchers used a neural network trained on existing IMU and video data to train their doppler sensor model, which achieved 75% accuracy with only a few seconds of its own training data. SMASH Lab's Rushil Khurana said, "People have already put so much time into labeling IMU data sets for everyday activities, and we can just piggyback on that to label new sensors such as a doppler."
|
Tool Allows Unprecedented Modeling of Magnetic Nanoparticles
NC State University News
Matt Shipman
March 11, 2022
North Carolina State University (NC State) researchers have designed a new computational tool that permits the modeling of multifunctional magnetic nanoparticles (MNPs) in unprecedented detail. NC State's Akhlak Ul-Mahmood said the tool accounts for MNPs' chemical interactions, adding that the team also developed "open-source software that the materials science community can use to implement it." The researchers demonstrated the tool's accuracy by modeling oleic acid ligand-functionalized magnetite nanoparticles, which Mahmood said predicted their behavior and characteristics in a manner consistent with what is known about them via experimental observation. The model also uncovered new insights about the MNPs' self-assembly behavior.
|
Linux Bug Gives Root on All Major Distros, Exploit Released
BleepingComputer
Lawrence Abrams
March 7, 2022
Security researcher Max Kellermann recently disclosed his discovery of the Dirty Pipe Linux bug, which lets local users obtain root privileges through publicly available exploits, and impacts Linux Kernel 5.8 and later iterations, even on Android devices. He released a proof-of-concept exploit that allows local users to inject their own data into sensitive read-only files, stripping restrictions or tweaking configurations to expand their access privileges. Kellermann alerted various Linux maintainers about Dirty Pipe beginning Feb. 20, and although it has been corrected in Linux kernels 5.16.11, 5.15.25, and 5.10.102, many servers still are running outdated kernels.
|
UCI Researchers Develop Hybrid Human-Machine Framework for Building Smarter AI
UCI News
March 7, 2022
University of California, Irvine (UCI) researchers unveiled a mathematical model that can improve the performance of artificial intelligence (AI) via a hybrid human-machine framework. "We show through empirical demonstrations as well as theoretical analyses that humans can improve the predictions of AI even when human accuracy is somewhat below [that of] the AI—and vice versa," said UCI's Mark Steyvers. "And this accuracy is higher than combining predictions from two individuals or two AI algorithms." The model was tested by having humans and algorithms separately classify distorted images of animals and everyday items, with humans ranking their confidence in the accuracy of each image identification, while the algorithms produced a continuous score. Confidence differed widely across images, and the hybrid model's performance topped that of either human or machine predictions.
|
Virginia Lawmakers OK Lifting Ban on Facial Technology Use
Associated Press
Denise Lavoie
March 10, 2022
Virginia lawmakers have passed legislation to lift a ban on facial recognition technology and to permit police to use it in certain scenarios, like helping to identify individuals under reasonable suspicion of committing crimes. The measure forbids the technology’s use for surveillance or monitoring. Sponsor Sen. Scott Surovell (D-VA) said the technology, when used appropriately, can be a valuable tool for solving and prosecuting crimes more efficiently. Some lawmakers opposed lifting the ban, raising issues about privacy, civil rights violations, and research indicating facial recognition software generates higher error rates when used to identify people of color. Some lawmakers said the new bill has "guardrails" to guarantee that law enforcement must comply with rules on when and how to use the technology.
|
|
