Welcome to the August 31, 2022, edition of ACM TechNews, providing timely information for IT professionals three times a week.

Please note: In observance of the U.S. Labor Day holiday, ACM TechNews will not be published on Friday, Sept. 2 and Monday, Sept. 5. Publication will resume Wednesday, Sept. 7.

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."
Researchers Bring Underwater Messaging App to Smartphones
Allen School News (University of Washington)
Kristin Osborne
August 29, 2022


The AquaApp mobile interface developed by University of Washington (UW) researchers facilitates underwater messaging using acoustic signals. "With AquaApp, we demonstrate underwater messaging using the speaker and microphone widely available on smartphones and watches," said UW's Tuochao Chen. AquaApp users can choose among 240 preset messages corresponding to hand signals employed by divers, with the 20 most-used signals displayed for easy access; they also can screen messages through categories such as directional indicators, environmental factors, and equipment status. AquaApp features an algorithm that optimizes the bitrate and acoustic frequencies of each transmission based on certain parameters in real time, while a networking protocol shares access to an underwater network.

Full Article

Researcher Sung Yun Jun checks the alignment of the transmitter, mounted 6 meters high on a mast. Streetlight Wi-Fi Hotspots: A Bright Idea
IEEE Spectrum
Charles Q. Choi
August 30, 2022


A new algorithm could help telecommunication companies determine the optimal deployment of Wi-Fi hotspots in streetlights. Researchers at the U.S. National Institute of Standards and Technology have created a model for tracking 60-gigahertz signals that account for variances in light pole height, with the algorithm analyzing transmission paths to include receiver sites. The researchers tested the model in Boulder, CO, with a stationary transmitter on a mast and a mobile receiver on a van, measuring signals at four-, six-, and nine-meter elevations. The model successfully predicted real-world measurements with an accuracy comparable to that of more complex methods.

Full Article
Optimizing Fluid Mixing with Machine Learning
Tokyo University of Science (Japan)
August 29, 2022


Researchers in Japan have proposed a machine learning-based approach for optimizing fluid mixing for laminar flows. The researchers used reinforcement learning (RL), in which intelligent agents perform actions in an environment to maximize the cumulative reward. The team addressed RL's inefficiencies in dealing with systems involving high-dimensional state spaces by describing fluid motion using only a single parameter. Researchers used the Markov decision process to formulate the RL algorithm, and the Tokyo University of Science's Masanobu Inubushi said the program "identified an effective flow control, which culminated in an exponentially fast mixing without any prior knowledge." The RL method also enabled effective transfer learning of the trained "mixer," significantly reducing its time and training cost.

Full Article

PetTrack uses a combination of sensors to provide the accurate, real-time indoor location of an animal. PetTrack Lets Owners Know Where Their Dog Is
Georgia Tech Research
Tess Malone
August 30, 2022


The PetTrack system developed by Georgia Institute of Technology (Georgia Tech) researchers uses a combination of sensors to tell pet owners where their animal is in real time. "PetTrack comprises two things: one is knowing the pet's indoor location and second is trying to understand their activity," said Georgia Tech's Ashutosh Dhekne. A device that can be attached to a pet’s collar features ultra-wideband radio wireless sensors to determine the location of the animal from up to 100 feet away, as well as accelerometers to determine whether it is standing, sitting, or at rest. Users can view this information through a smartphone application. Dhekne said, "Together, combining where the pet is and what the orientation of the pet is, we can create a summary map of where the pet has been during the day and what activity the pet was doing."

Full Article
Hackers Hide Malware in James Webb Telescope Images
BleepingComputer
Bill Toulas
August 30, 2022


Hackers have launched a campaign dubbed GO#WEBBFUSCATOR by threat analysts to spread malware via phishing emails, malicious documents, and James Webb telescope images. Malefactors drop payloads currently not labeled malicious by antivirus engines on the VirusTotal scanning platform, and infiltration begins with a phishing email with an attached document that downloads a template file. The file features an obfuscated VBS macro that auto-executes if macros are enabled in the Office suite, then downloads, decodes, and launches a JPG image of a galactic cluster. Security analytics company Securonix says the malware creates a Domain Name System link to the command and control (C2) server, and sends encrypted queries that "are read in and unencrypted on the C2 server, thus revealing its original contents." Securonix has furnished indicators of compromise, including network and host-based indicators.

Full Article

The artificial hand contains artificial pneumatic muscles consisting of 3D-printed structures that can extend and contract as required. 3D Artificial Pneumatic Muscles for Future 'Makers'
IIT - Italian Institute of Technology
August 30, 2022


Researchers at Italy's Istituto Italiano di Tecnologia (the Italian Institute of Technology, IIT) and the Scuola Superiore Sant'Anna (the Sant'Anna School of Advanced Studies) have developed artificial pneumatic muscles that can be produced using three-dimensional (3D) printers. The researchers demonstrated a pneumatic hand made of 18 GRACE (GeometRy-based Actuators able to Contract and Elongate) actuators shaped like a spindle with pleats, allowing them to expand, extend, and contract with little pressure. Said IIT's Corrado De Pascali, "Their size is limited purely by the manufacturing technology used. They can be built in different sizes, and we can vary their performance, both in terms of deformation and strength, and manufacture them using various materials and technology, even already built into the structures to be fabricated."

Full Article
From Bus Routes to Gutters, Tech-Savvy Youth Map Mali's Capital
Yahoo! News
August 29, 2022


Volunteers with OpenStreetMap Mali are collecting data on the infrastructure in Mali's capital, Bamako, including its public minibus routes, household waste collection points, basic social services, and roadside drains. A group of young, tech-savvy Malians are using a smartphone mapping app to gather data for the local branch of the free geographic database OpenStreetMap, which is used by Google Maps and other sites. Adama Konate, Bamako’s deputy mayor in charge of sanitation, said, "We only had basic knowledge before this project. Now we know that this place needs drainage, and that place needs a rubbish dump."

Full Article
AI Can Learn the Patterns of Human Language
MIT News
Adam Zewe
August 30, 2022


Researchers at Massachusetts Institute of Technology, Cornell University, and McGill University developed an artificial intelligence model that can learn the rules and patterns of human languages automatically, without specific human guidance. The model was trained and tested on problems from linguistic textbooks in 58 different languages that involved word-form changes. The researchers observed that the model could determine a correct set of rules to describe the word-form changes for 60% of the problems. Said Cornell's Kevin Ellis, "One of the things that was most surprising is that we could learn across languages, but it didn't seem to make a huge difference. That suggests two things. Maybe we need better methods for learning across problems. And maybe, if we can't come up with those methods, this work can help us probe different ideas we have about what knowledge to share across problems."

Full Article

Artificial intelligence can help researchers create photorealistic images from three-dimensional scenery, as improved realism can help improve testing of autonomous driving. Driving Simulations That Look More Lifelike
Ohio State News
August 29, 2022


Ohio State University (OSU) researchers trained a Generative Adversarial Network (GAN) to create random photorealistic three-dimensional scenery for driving simulations. The researchers based their simulator on the open-source CARLA tool, then applied a GAN-based image synthesizer to render and integrate background elements with more conventionally rendered objects. The researchers determined that combining foreground objects separately from background scenery resulted in an image that was more lifelike. OSU's Ekim Yurtsever said, "We can never actually replace real-world testing, but if we can make simulations a little bit better, we can get better insight on how we can improve autonomous driving systems and how we interact with them."

Full Article

French tax authorities are using AI software to find thousands of undeclared private swimming pools. French Tax Officials Use AI to Spot 20,000 Undeclared Pools
The Guardian (U.K.)
Kim Willsher
August 29, 2022


French tax officials used artificial intelligence (AI) from Google and Capgemini to identify over 20,0000 undeclared private swimming pools, constituting about €10 million (over $10 million) in tax receipts. The AI system can spot pools in aerial images, and cross-checks them with land registry databases to determine which may have not been reported. The tax authorities' technical team said it cannot yet establish whether a rectangular configuration on an aerial image is an extension or a tent, terrace, or tarpaulin placed on the ground. Antoine Magnant, France's deputy director general of public finances, said tests of the system are only in their second stage.

Full Article

The machine learning algorithm allows a robot dog like this one to learn to walk on unfamiliar and hard-to-master terrains in about 20 minutes. Robot Dog Learns to Walk Tough Terrain in 20 Minutes
New Scientist
Alex Wilkins
August 26, 2022


Researchers at the University of California, Berkeley (UC Berkeley) developed a machine learning algorithm that enabled a robot dog to learn to navigate difficult terrain in only 20 minutes. The Q-learning algorithm does not need a model of the target terrain. As a result, said UC Berkeley's Sergey Levine, "We don't need to understand how the physics of an environment actually works, we just put the robot into an environment and turn it on." The algorithm teaches the robot by rewarding it for each successful action until reaching its ultimate goal. The researchers demonstrated that the robot was able to walk on terrains it had not previously encountered, including grass, a layer of bark, a memory foam mattress, and a hiking trail, after about 20 minutes of training on each.

Full Article
Model Examining Materials at Mesoscale May Lead to Next-Gen Devices
Penn State News
Matthew Carroll
August 30, 2022


A new phase-field computational model developed by Pennsylvania State University (Penn State) researchers can characterize the dynamics of simultaneous structural and electronic processes in functional and quantum materials at the mesoscale (between the atomic scale and what is observable by the human eye). The model could serve as a link between atomic-scale quantum mechanical calculations and devices for enabling next-generation quantum technologies. "This model now lets us look at the frequency dependence of these responses and see how the structure has actually evolved inside the material and how that connects to the properties," explained Penn State's Long Qing Chen.

Full Article
Algorithm Optimizes Antibiotic Use in Settings with Poor Hygiene, Sanitation
University of Florida Health
Doug Bennett
August 29, 2022


An international group of scientists developed an artificial intelligence-based tool that algorithmically calculates the likelihood that a case of diarrhea is caused by a virus alone, in order to limit the misuse of antibiotics. The Diarrheal Etiology Prediction (DEP) algorithm mines datasets that include clinical history, patient-specific symptoms, and local weather data. The researchers used it to validate their hypothesis that it would help inform physicians' decisions to prescribe antibiotics to children with diarrhea. The project involved 30 doctors and the health records of nearly 1,000 patients between two months and five years old at seven sites in Bangladesh and Mali.

Full Article
September 2022 Issue of Communications of the ACM
 
ACM Insurance for Members
 

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]