Welcome to the October 2, 2023, 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."
|
|
|
Gamers Help Highlight Disparities in Algorithm Data
Cornell Chronicle
Louis DiPietro
September 29, 2023
Researchers from Cornell University, Microsoft-owned gaming brand Xbox, and Microsoft Research tapped an international pool of 5,174 gamers to improve predictive models' ability to provide personalized gaming recommendations. The gamers labeled massive datasets of gaming titles that the researchers fed to the models. The gamers applied certain game labels consistently across countries; others were used inconsistently, and the researchers used computational methods to explain the inconsistencies' cultural, translational, and linguistic contributors. They created two predictive models respectively based on survey data from globally representative gamers and data from U.S. gamers exclusively. The first model improved prediction 8% for gamers everywhere versus the second model.
|
AI Deepfakes Spread Disinformation in Slovak Elections
Bloomberg
Olivia Solon
September 29, 2023
Disinformation was spread over social media in the run-up to the Slovak elections that took place over the weekend, with videos featuring artificial intelligence (AI)-produced deepfake voices. One video shows a conversation in which Slovakian progressive party leader Michal Simecka appears to discuss vote-buying from the Roma minority with a journalist, which experts deemed synthesized by an AI tool trained on samples of the speakers' voices. Technological democracy research group Reset's Rolf Fredheim said, "With the examples from the Slovak election, there's every reason to think that professional manipulators are looking at these tools to create effects and distribute them in a coordinated way."
*May Require Paid Registration
|
Dark Patterns Found in Japanese Mobile Apps
Tokyo Tech News (Japan)
September 29, 2023
Scientists at Japan's Tokyo Institute of Technology (Tokyo Tech) have uncovered dark patterns (DPs), also known as deceptive user interface elements, designed into popular Japanese mobile applications. Most of the examined apps contained DPs, with each app featuring approximately 3.9 deceptive design elements on average. The researchers assigned these patterns to six classes, including a new class—"Linguistic Dead-Ends"—consisting of the "Untranslation" and "Alphabet Soup" subclasses. Tokyo Tech's Katie Seaborn explained, "Linguistic Dead-Ends refer to language and symbol use that prevents a user from understanding crucial functionality in an app, even while the rest of the app is in perfect Japanese."
*May Require Paid Registration
|
Ant Brains Inspire Robots to Find Their Way
Interesting Engineering
Rizwan Choudhury
September 27, 2023
A team led by researchers at the U.K.'s University of Edinburgh developed an artificial neural network based on the brain structure of ants to help robots navigate dense, plant-filled landscapes and other complex natural environments. The design of the neural network mimics the mushroom-like neuron structures in the brains of ants, which allow them to identify visual patterns and retain spatiotemporal memories, helping them learn and navigate routes in visually repetitive surroundings. The researchers equipped a terrestrial robot with a bioinspired event camera that captures visual sequences along routes in natural outdoor environments. They then applied a neural algorithm for spatiotemporal memory based on the mushroom body circuit and encoded memory in a spiking neural network operating on a low-power neuromorphic computer. The neural model was found to outperform SeqSLAM, an existing route learning model.
|
Researchers Fabricate Chip-Based Optical Resonators with Record Low UV Losses
Optica
September 26, 2023
Researchers at Yale University and semiconductor materials developer Entegris have manufactured chip-based photonic resonators that exhibit unprecedented low loss at ultraviolet (UV) wavelengths. The researchers etched alumina into a rib waveguide, producing ring resonators with a 400-micron radius. They found etching to a depth of more than 80 nanometers in a 400-nanometer-thick alumina layer could suppress radiation loss to less than 0.06 decibel/centimeter at 488.5 nanometers and less than 0.001 decibel/centimeter at 390 nanometers. Said Yale's Chengxing He, “Compared to the better-established fields like telecom photonics and visible photonics, UV photonics is less explored even though UV wavelengths are needed to access certain atomic transitions in atom/ion-based quantum computing and to excite certain fluorescent molecules for biochemical sensing."
|
Adding Nanofridges to Quantum Computers Could Make Them Run Faster
New Scientist
Karmela Padavic-Callaghan
October 2, 2023
A multi-institutional team of researchers led by Teruaki Yoshioka at Japan's Tokyo University of Science has shortened the quantum bit (qubit) reset process through the use of a nano-sized quantum circuit refrigerator (QCR). The researchers constructed a qubit on a chip before adding a small resonator and the aluminum-copper QCR. When the qubit's state was too energetic to perform a calculation, the researchers could pass some of its energy to the resonator, which would render it as a light particle. An electron in the QCR captured that particle, cooling the qubit and reducing its reset time to about 100 nanoseconds.
|
Wearable Device Data Links Reduced Sleep, Activity in Pregnancy to Premature Birth Risk
Stanford Medicine News Center
Erin Digitale
September 28, 2023
Scientists at the Stanford School of Medicine (Stanford Medicine) and Washington University in St. Louis associated a lack of sleep and reduced physical activity during pregnancy with the risk of premature birth. Stanford Medicine's Nima Aghaeepour said, "We showed that an artificial intelligence algorithm can build a 'clock' of physical activity and sleep during pregnancy and can tell how far along a patient's pregnancy is." The researchers mined information from actigraphy devices (used to assess cycles of activity and rest) worn by more than 1,000 women during pregnancy, then used a machine learning algorithm to identify changes in sleep and physical activity. They found women whose patterns changed faster relative to their pregnancies' progress tended to deliver their babies early.
|
Chrome Zero-Day Sends the Internet into New Chapter of Groundhog Day
Ars Technica
Dan Goodin
September 28, 2023
A critical zero-day vulnerability reported on Sept. 27 by Google that affects its Chrome browser, as well as Mozilla's Firefox browser, is similar to one reported on Sept. 11. Both vulnerabilities reside in a code library for processing media files, particularly in the VP8 format. The most recent zero-day reported by Google was exploited in the code library libvpx and reportedly applies to video encoding. The previous zero-day was exploited in the code library libwebp and applied to encoding and decoding. Analygence's Will Dorman said, "The [vulnerability] is in VP8 encoding, so if something uses libvpx only for decoding, they have nothing to worry about." Dorman added that "Firefox, Chrome (and Chromium-based) browsers, plus other things that expose VP8 encoding capabilities from libvpx to JavaScript (i.e. Web browsers), seem to be at risk." Patches have been made available for Chrome and Firefox.
|
Facing Up to Bird Ecology
Flinders University (Australia)
October 2, 2023
Researchers at Austria's University of Vienna (UNIVIE) and Australia's Flinders University used automated facial recognition software to study the social behavior of Greylag Geese in Europe. The researchers designed the software to assess a goose face's visual uniqueness based on the similarity between images of goose bills. They programmed the software to evaluate each bird's response to images of themselves and tested each goose face with life-sized two-dimensional images of flock members. The researchers found the software could correctly assign each goose face within a photo library containing 6,000 possible matches with 97% accuracy, then learned geese behave differently when presented with individual images in a grassy field, said UNIVIE's Sonia Kleindorfer.
|
Decoding the Complexity of Alzheimer's Disease
MIT News
Anne Trafton
September 28, 2023
Multi-institutional teams of researchers led by the Massachusetts Institute of Technology (MIT) analyzed the genomic, epigenomic, and transcriptomic changes occurring across all cell types in the brains of Alzheimer's patients in the hope of finding new therapeutic targets. Said MIT's Manolis Kellis, "What we set out to do was blend together our computational and our biological expertise and take an unbiased look at Alzheimer's at an unprecedented scale across hundreds of individuals." One research team defined gene-expression disruption as the disease evolves via single-cell RNA-sequencing, using 54 types of cells identified in 427 postmortem brain samples. Another team used the ATAC-Seq method plus single-cell RNA sequencing to monitor cells' epigenomic changes among 48 controls and 44 people with early or late-stage Alzheimer's. Other investigations concentrated on microglia and DNA damage's contributions to Alzheimer's.
|
Multi-Purpose Robot Changes Shape for Different Uses
University of Bristol News (U.K.)
September 29, 2023
A tetrahedron-shaped robot developed by researchers at the U.K.'s University of Bristol features flexible piping that allows it to alter its shape, size, and movements based on its intended use. Tetraflex can navigate small spaces and difficult terrain and can transport fragile items by enveloping them with its body. The robot is made of soft struts composed of an airtight rubber bellow connected by rigid nodes. Changing the air pressure in the bellow controls the shape and size of the robot, and altering the length of the struts in a particular sequence produces different types of motion.
|
|
