Welcome to the March 21, 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."

Marina Ovsyannikova protests Russian invasion of Ukraine on set of Russia's state TV flagship program Computer Programmers Take Aim at Russia's Propaganda Wall
The Washington Post
Drew Harwell
March 17, 2022


Computer programmers and volunteer "information warriors" are attempting to counter Russian propaganda and information suppression concerning the Ukraine invasion. A Website built by the squad303 coder group shows a randomly selected Russian citizen's email address and phone or WhatsApp number, and provides a pre-written message visitors can send to engage in a dialogue. A Polish programmer said he works with more than 100 volunteers from the U.S., Estonia, France, Germany, and more, divided into teams focused on software development, cyberdefense, social media, and a help desk to onboard new messengers. Western social media companies and media outlets also have started helping Russians bypass government censorship by using Tor software, which directs online traffic through a scattered network of servers, neutralizing Russia's Website blockade. Market research data indicates virtual private network applications, which enable Russians to access otherwise-banned sites, have been downloaded millions of times on the Apple and Google app stores.

Full Article

Examples of the generative model's transformation methods. When It Comes to AI, Can We Ditch the Datasets?
MIT News
Adam Zewe
March 15, 2022


Massachusetts Institute of Technology (MIT) researchers have demonstrated the use of a generative machine-learning model to produce synthetic data, based on real data, to train another model for image classification. Researchers showed the generative model millions of images containing objects in a specific class, after which it learned those objects' appearance in order to generate similar objects. MIT's Ali Jahanian said generative models also learn how to transform underlying training data, and connecting a pretrained generative model to a contrastive learning model enabled both models to work together automatically. The results show that a contrastive representation learning model trained only on synthetic data can learn visual representations that rival or top those learned from real data. In analyzing how the number of samples influenced the model's performance, researchers determined that, in some cases, generating larger numbers of unique samples facilitated additional enhancements.

Full Article
Beware of QR Code Scams
The Wall Street Journal
Heidi Mitchell
March 19, 2022


Security researchers warn of the growing threat of fraudulent quick response (QR) codes, including some affixed to parking meters in Texas cities that tricked drivers into entering their credit-card data at a bogus Website. Although the Better Business Bureau's Scam Tracker site lists just 46 QR code-related attacks in the U.S. since March 2020, link-management service Bit.ly has observed a 750% increase in QR-code downloads since then. Most smartphones "just read the code and open the link without ensuring that it is safe or that it is, in fact, what it says it is," said Justin Fier at artificial intelligence cybersecurity firm Darktrace. Skilled attackers also can use a QR code to send users to a spoof site, then hand over the information they enter to the genuine site. Symantec's Eric Chien suggests either avoiding QR codes that are stuck on devices or installing QR-code scanner applications.

Full Article
*May Require Paid Registration
Decoding Brain Signals to Control Robotic Arm
KAIST (South Korea)
March 18, 2022


A neural signal decoder system developed by researchers at the Korea Advanced Institute of Science and Technology (KAIST) can control a robotic arm and eventually could be used in real-time brain-machine interfaces. The researchers' method for decoding electrocorticography (ECoG) neural signals relies on an "echo-state network" and a mathematical probability model. With ECoG, used mainly to detect sources of epileptic seizures, the electrodes may not be in the optimal areas of the brain to detect sensory and movement signals. In the study of four individuals with epilepsy, the percentage of electrodes placed in these regions of the brain ranged from 22% to 44%. However, the decoding system was able to classify arm movements in both real and virtual tasks in 24 directions, and a computer simulation demonstrated that it could control a robotic arm. Researchers found that the system was at least five times more accurate than chance.

Full Article

SFU archaeology PhD student and interim director of SFU’s Museum of Archaeology and Ethnology, Rob Rondeau. Archaeologists Collaborate on Tech that IDs Ancient Hunter-Gatherer Sites
Simon Fraser University (Canada)
Melissa Shaw
March 17, 2022


Archaeologists at Germany's Max Planck Institute and Canada's Simon Fraser University (SFU) are using a computer predictive model to identify sites used by nomadic hunter-gatherers. The Locally-Adaptive Model of Archaeological Potential (LAMAP) uses information from landscape data to calculate the archaeological potential of land that archaeologists have not scrutinized. SFU's Rob Rondeau and Max Planck's Chris Carleton, who created LAMAP, have partnered to find ancient hunter-gatherer campsites on submerged land off the coast of British Columbia. They tested LAMAP on Alaska's Tanana Valley, training the model with data about 90 known sites randomly chosen from Alaska's Heritage Database. Regions that LAMAP forecast as high potential contained many of the remaining sites in the database.

Full Article
Time Releases Full Magazine Issue as NFT on the Blockchain
Time
March 18, 2022


Time magazine will publish the first fully decentralized magazine issue, available on March 23 as a non-fungible token (NFT) on the blockchain. Created in partnership with LITDAO, a Web3 cultural currency and NFT project, the issue will be hosted through a decentralized protocol, with readers accessing the magazine through an interactive NFT. With support from the global Internet finance firm Circle, the issue, which will feature a cover story on Ethereum's Vitalik Buterin, will be airdropped to certain TIMEPiece and genesis LIT community wallet holders. "As Time continues to push the boundaries as to what is possible within the Web3 ecosystem, producing the first-ever full magazine on the blockchain seemed like a natural extension for our brand, and we knew this issue, in particular, would be cherished by our community," said Time's Keith A. Grossman.

Full Article
Nearly All Websites May Breach GDPR Legislation around Data Usage
New Scientist
Chris Stokel-Walker
March 18, 2022


An analysis of nearly 29,398 websites by researchers at Switzerland's ETH Zurich found 94.7% potentially violated Europe's General Data Protection Regulation (GDPR). Since 2018, the GDPR has required sites to have cookie bars or banners that pop up when users first visit, explaining why cookies are collected, defining their usage, and requiring users to consent to their data's storage. Researchers evaluated whether the banner text accurately represented the cookies being collected and if user consent impacted the cookies saved. They developed a Web browser extension that uses machine learning to categorize cookies as either essential, functional, analytical, or ad-related. About 70% of the sites activated cookies before a user had consented to data storage, and about 20% activated cookies the user had declined.

Full Article

BirdBot robot. BirdBot Is Energy-Efficient Thanks to Nature as a Model
Max Planck Institute for Intelligent Systems
March 16, 2022


Scientists at Germany's Max Planck Institute for Intelligent Systems (MPI-IS) and the University of California, Irvine have built BirdBot, an energy-efficient robot leg modeled after the leg of a flightless bird. BirdBot uses a foot-leg coupling via a network of muscles and tendons extended across multiple joints, requiring fewer motors than other legged robots, which could, in theory, be scaled up. Researchers walked the device on a treadmill to observe its foot folding and unfolding. "Our results demonstrating this mechanism in a robot lead us to believe that similar efficiency benefits also hold true for birds," said Alexander Badri-Spröwitz from MPI-IS.

Full Article

John Deere autonomous tractor plows a field, without a driver. Tired of Waiting for Driverless Vehicles? Head to a Farm
Associated Press
Scott McFetridge
March 16, 2022


Driverless vehicles are more abundant on farms than city streets, with John Deere to start manufacturing autonomous tractors this fall after more than 10 years in development. The company intends to run the tractors on 10 to 50 farms by fall, before expanding to more farms in the coming years. Carnegie Mellon University's Raj Rajkumar said autonomous tractors have no vehicles, pedestrians, or intricacies of urban systems to deal with, and they can employ consistent global-positioning system data. Farmers can hitch a plow behind the driverless tractor, start it with a swipe of a smartphone, and then leave it to travel the field on its own. The machine has six pairs of cameras that can provide a 360-degree image, and computer algorithms help it to navigate and stop before unfamiliar obstacles.

Full Article
Using AI-powered Apps May Speed Up Stroke Treatment
The Pittsburgh Post-Gazette
Anya Sostek
March 20, 2022


Pittsburgh hospitals are using artificial intelligence (AI)-powered applications to identify patients at risk for stroke and to alert doctors. The University of Pittsburgh Medical Center (UPMC) is using the Viz.ai app to read computed tomography (CT) scans of patients and send smartphone alerts to doctors and nurses if it finds abnormalities. Meanwhile, Allegheny Health Network employs the RapidAI app for similar functions; after its introduction, the time between CT scans and artery unblocking procedures decreased from an average of 93 minutes to 68 minutes. UPMC's Raul Nogueira said the AI does not replace clinicians, but it can flag things faster and perhaps identify abnormalities a doctor might overlook. He added that Viz.ai and its underlying AI have been in development for roughly five years, while UPMC will expand its deployment from three hospitals in January to an additional two dozen through June.

Full Article

Yu Harabuchi, Wataru Matsuoka, and Satoshi Maeda (left to right) of the research team at the Institute for Chemical Reaction Design and Discovery (ICReDD) at Hokkaido University. Chemical Reaction Design Goes Virtual
Hokkaido University (Japan)
March 14, 2022


A virtual ligand-assisted (VLA) screening method developed by researchers at Japan's Hokkaido University could reduce time spent on trial and error during transition metal catalyst development. The VLA screening method surveys a wide range of values for different properties to determine the most promising features of ligands, which are molecules bonded to the central metal atom of a catalyst. Virtual ligands mimic the presence of real ligands, but rather than being described by numerous individual constituent atoms, they are described using only their steric (space-filling) and electronic properties. Researchers then generate a contour map detailing the combination of steric and electronic effects a ligand should have to best catalyze a specific reaction, so chemists only must test real ligands meeting this criteria. "As the VLA screening can be conducted in silico, it would save a lot of time and resources in the lab," said Hokkaido's Satoshi Maeda.

Full Article
Unix Rootkit Used to Steal ATM Banking Data
BleepingComputer
Bill Toulas
March 17, 2022


Researchers at the cybersecurity firm Mandiant found that the LightBasin hacking group is using a previously unknown Unix rootkit to steal ATM banking data and make unauthorized cash withdrawals from ATM terminals at several banks. The rootkit, a Unix kernel module called "Caketap," affects servers running the Oracle Solaris operating system, hiding network connections, processes, and files while installing several hooks into system functions to receive remote commands and configurations. Caketap intercepts messages sent to the Payment Hardware Security Module (HSM), used by the banking industry to verify bank card information, to stop verification messages that match fraudulent bank cards and instead generate a valid response. It also internally saves valid messages that match non-fraudulent primary account numbers and sends them to the HSM to avoid impacting routine customer transactions and implant operations.

Full Article
ACM Transactions on Evolutionary Learning and Optimization
 
ACM Distinguished Speakers Program
 

Association for Computing Machinery

1601 Broadway, 10th Floor
New York, NY 10019-7434
1-800-342-6626
(U.S./Canada)



ACM Media Sales

If you are interested in advertising in ACM TechNews or other ACM publications, please contact ACM Media Sales or (212) 626-0686, or visit ACM Media for more information.

To submit feedback about ACM TechNews, contact: [email protected]