Welcome to the July 15, 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."
|
|
A Robot Learns to Imagine Itself
Columbia Engineering News Holly Evarts July 13, 2022
Columbia Engineering researchers have created a robot that can learn a model of its body without human intervention. The robot produced a kinematic model of itself to plan motion, achieve goals, avoid obstacles, and identify and compensate for bodily damage. The researchers installed a robotic arm within a ring of streaming video cameras, enabling the device to observe itself while undulating, and to learn how its body moved in response to motor commands. Columbia's Hod Lipson said a self-modeling robot or creature "can function better in the world, it can make better decisions, and it has an evolutionary advantage."
|
Log4j Software Flaw 'Endemic,' Cyber Safety Panel Says
Associated Press Alan Suderman July 14, 2022
The Cyber Safety Review Board said the Log4j software vulnerability discovered last year is "endemic," and could constitute a security risk for another decade. Log4j enables Internet-based hackers to hijack a broad range of systems; the first indications of its exploitation appeared in Microsoft's online game Minecraft. Log4j logs user activity on computers, and is widely employed by commercial software developers. Although the review board has found no signs of "significant" Log4j attacks on critical infrastructure systems, it said future attacks are likely. To alleviate the potential fallout of such attacks, the board recommended universities and community colleges make cybersecurity training mandatory for obtaining computer science degrees and certifications.
|
The Webb Space Telescope's Profound Data Challenges
IEEE Spectrum Michael Koziol July 8, 2022
The James Webb Space Telescope (JWST)'s data-collecting operations present a number of challenges, including dependence on a reliable communications subsystem. The spacecraft is sending up to 57 gigabytes (GB) of data per day back to Earth on a 25.9-gigahertz channel at up to 28 megabits per second. Data recorded by the JWST's scientific instruments is stored in the spacecraft's 68-GB solid-state drive, which can collect data for about 24 hours before reaching its limit. Only after the spacecraft receives confirmation of a data file's receipt will it delete its onboard copy of the data to clear space. The telescope will stay connected to Earth through the Deep Space Network, sharing limited antenna time with other deep-space missions.
|
Researchers Find Missing Photonic Link to Enable All-Silicon Quantum Internet
Simon Fraser University (Canada) Erin Brown-John July 13, 2022
Researchers at Canada's Simon Fraser University (SFU) have found that T centers, a luminescent impurity in silicon, can support a photonic link between quantum bits (qubits). SFU's Stephanie Simmons said, "This work is the first measurement of single T centers in isolation, and actually, the first measurement of any single spin in silicon to be performed with only optical measurements." Simmons called T centers an ideal component for scalable, distributed quantum computers, "because they can handle the processing and the communications together, rather than needing to interface two different quantum technologies, one for processing and one for communications." Silicon qubits interact by emitting light within the same band used by datacenters and fiber networks, which Simmons said yields "the same benefits for connecting the millions of qubits needed for quantum computing."
|
You've Been Served Via NFT: Court Gives OK to Sue on Blockchain
Bloomberg Katharine Gemmell July 13, 2022
A U.K. court ruling allows legal documents to be served over the blockchain ledger via nonfungible tokens (NFTs). The case was filed by Fabrizio D'Aloia, founder of an online gambling company, against Binance Holdings and other cryptocurrency exchanges after his crypto assets were fraudulently cloned. The exchanges also were deemed responsible for ensuring stolen crypto is not moved or removed from their systems. Legal experts at the law firm Giambrone & Partners LLP said the ruling will enable crypto fraud victims to file suit against unknown fraudsters in the U.K. The lawsuit documents will be airdropped via NFT into two wallets originally used by D'Aloia and later stolen. A similar decision was issued in June by a U.S. court.
*May Require Paid Registration
|
Computer Model Predicts Whether a Pesticide Will Harm Bees
Oregon State University News Steve Lundeberg July 13, 2022
Oregon State University (OSU) researchers trained a machine learning model to forecast whether new herbicides, fungicides, or insecticides would harm honey bees. The researchers used honey bee toxicity data from pesticide exposure experiments involving roughly 400 pesticide molecules to teach an algorithm to predict a new pesticide molecule's toxicity. OSU's Ping Yang said the model equates pesticide molecules with a series of random walks (mathematical equivalents of meandering paths) on their molecular graphs. Yang explained, "The algorithm declares two molecules similar if they share many walks with the same sequence of atoms and bonds. Our model serves as a surrogate for a bee toxicity experiment and can be used to quickly screen proposed pesticide molecules for their toxicity."
|
Teaching AI to Ask Clinical Questions
MIT Computer Science and Artificial Intelligence Laboratory Adam Zewe July 14, 2022
Massachusetts Institute of Technology (MIT) scientists and medical experts have compiled a dataset of 2,000-plus questions physicians ask when reviewing electronic health records (EHRs). A machine learning (ML) model trained on this dataset asked authentic questions, versus actual questions from medical experts, over 60% of the time. In contrast, models trained to recover answers to clinical questions using publicly available datasets could only retrieve about 25% of answers to physician-generated questions. MIT's Peter Szolovits said, "The value of this work is in carefully collecting questions asked by clinicians about patient cases, from which we are able to develop methods that use these data and general language models to ask further plausible questions."
|
Monitoring City Mobility from the Skies
EPFL News (Switzerland) July 13, 2022
Researchers from École Polytechnique Fédérale de Lausanne (EPFL) in Switzerland used aerial drone fleets to measure city traffic. The researchers employed the drone-compiled traffic data to develop an algorithmic method for identifying congestion sources and offering recommendations to relieve multimodal problems. They deployed the CityDronics system in May to track and assess traffic in Nairobi, Kenya, which presented novel challenges like heavy congestion and heightened sensitivity to data confidentiality. Ten drones flew over the city along two avenues leading to the central business district during peak hours, amassing four days of data to be analyzed for traffic patterns. EPFL's Manos Barmpounakis said, "Our goal is not to monitor traffic, but to find the causes of congestion and provide solutions based on facts."
|
Videogame Prescribed by Doctors to Treat ADHD
BBC News Kitti Palmai July 11, 2022
A computer game called EndeavorRx is giving attention deficit hyperactivity disorder (ADHD) patients hope, after obtaining approval from the U.S. Food and Drug Administration to treat the condition in children. Researchers at technology company Akili developed EndeavorRx to stimulate and improve brain regions critical in attention function by training children to multitask and ignore distractions. An algorithm measures user performance and tailors the game's difficulty in real time, and U.S. doctors are prescribing the game to ADHD patients. U.K. application Thymia also is using computer games to help medical professionals identify and diagnose mental health issues. Both Akili and Thymia describe their apps as supplements to, and not a replacement for, physician-directed monitoring and therapy.
|
Safer Web Surfing with Method for Detecting Malicious Modes
SPIE Newsroom July 12, 2022
Scientists at South Korea's Far East and Namseoul universities have proposed a new technique for screening Websites for malicious codes by identifying and analyzing common distribution patterns. The researchers first "crawled" through 500 harmful sites to find such patterns, then focused on the programming methods and scripts used in those malicious codes. They added up how many times each method was used in malicious sites, and devised an equation to ascertain a given site's risk score. The technique is exceptionally accurate and fast, and Namseoul's Won-shik Na said its ability to identify malicious Websites from script patterns means the algorithm's complexity and memory cost is low. The approach also could identify zero-day attacks.
|
A Health Monitoring Wearable That Operates Without a Battery
University of California, Irvine July 12, 2022
University of California, Irvine (UCI) researchers have developed a self-powered, wireless wearable device that can monitor a user's pulse. The researchers indicated that a change in the sensor circuitry would allow for the monitoring of heart rate, body temperature, blood pressure, and other vital signs as well. The wristband features triboelectric nanogenerators (TENGs) that convert mechanical thumping or pressure into electricity, to power the device so it can display the user's pulse rate on an LED display. It also features near-field communication technology to allow for wireless power and data exchange between the wristband and a nearby mobile device. Said UCI's Rahim Esfandyar-Pour, the wristband "enables continuous, battery-free, wireless, and on-demand health monitoring anytime and anywhere. It's made with low-cost and flexible materials and can be tailored to meet a variety of wearable bioelectronic sensors' requirements."
|
How ML Can Identify Gun Buyers at Risk of Suicide
The Hill Gianna Melillo July 14, 2022
Research by University of California, Davis (UC Davis) scientists suggests machine learning could use handgun purchasing data to identify gun purchasers at risk of suicide. The researchers analyzed data from the California's Dealer's Record of Sale database, applying random forest classification to assess the data's ability to predict those at an elevated suicide risk within a year of a gun purchase. They estimated a 0.07% rate of suicide for gun buyers within that year, and identified 41 risk factors for firearm suicide, including older age, being a first-time purchaser, white race, living close to the seller, and the purchase of a revolver. Said UC Davis' Hannah S. Laqueur, "This study contributes to the growing evidence that computational methods can aid in the identification of high-risk groups and the development of targeted interventions."
|
Lime Has Built Its Own Camera-Based Sidewalks Detection Technology
TechCrunch Rebecca Bellan July 13, 2022
Shared micromobility company Lime unveiled plans to monitor scooter riders with cameras and computer vision technology to verify they are not riding on sidewalks. The startup will test its advanced rider assistance system on hundreds of scooters in San Francisco and Chicago starting next month, and hopes to add Paris and three other cities by year's end. The first iteration of Lime Vision is a retrofittable, waterproof unit that attaches to the scooter's neck containing the camera, an artificial intelligence chip, and a central processing unit. Lime's Joe Kraus said the computer vision model computes in real time, adding that he expects computer vision will be a more affordable solution for such operations than global positioning system augmentation, while also offering more relevant use cases.
|
|