Welcome to the March 8, 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."
|
|
New Factsheet Highlights Gender Disparities in Innovation, Technology
UNESCO March 7, 2023
On International Women’s Day, the United Nations Educational, Scientific, and Cultural Organization (UNESCO) released its latest #HerEducationOurFuture factsheet on gender equality in education, which cites persistent gender inequity in innovation and technology. Women's underrepresentation in science, technology, engineering, and math careers remains, with women constituting just 31% of research and development positions in science and only 24% of leadership positions in technology. Globally in 2018, 40% of computer science graduates were women. In 2019, in 61 out of 115 countries, fewer than 30% of computer science graduates were women. Only 22% of professionals in artificial intelligence (AI) and just 18% of authors at leading AI conferences are female.
|
FBI, Pentagon Helped Research Facial Recognition for Street Cameras, Drones
The Washington Post Drew Harwell March 7, 2023
Internal documents indicate the U.S. Federal Bureau of Investigation (FBI) and Department of Defense (DoD) actively participated in research and development of facial recognition software for identifying individuals observed by street cameras and aerial drones. The documents were disclosed in response to ongoing Freedom of Information Act litigation by the American Civil Liberties Union against the FBI. Many concern DoD’s Intelligence Advanced Research Projects Agency-funded Janus program, in which FBI scientists collaborated with computer-vision experts and program leaders on software that would accurately process "truly unconstrained face imagery" recorded by public surveillance cameras. The refined system was bundled into the Horus search tool released to the Pentagon's Combating Terrorism Technical Support Office.
*May Require Paid Registration
|
Could the Next Blockbuster Drug Be Lab-Rat Free?
The New York Times Emily Anthes March 7, 2023
Drugmakers hope to forgo animal testing and save time and labor collecting safety and effectiveness data on medications with new tools like organs on chips and computer models. Certain computer models can predict the likely toxicity or metabolization rate of a compound with specific chemical properties. Such models can be tweaked to represent different types of patients, allowing developers to test whether drugs effective for young adults would be safe and effective in older adults, for example. Virtual cells also have potential, with researchers able to simulate individual human heart cells using "a set of equations that describe everything that's going on in the cell," according to Elisa Passini at the U.K.'s National Center for the Replacement, Refinement, and Reduction of Animals in Research.
*May Require Paid Registration
|
A 4D Printer for Smart Materials with Magneto-, Electro-Mechanical Properties
Universidad Carlos III de Madrid (Spain) March 6, 2023
Engineers at Spain's Universidad Carlos III de Madrid (UC3M) have developed software and hardware for a 4D printer that can produce soft multifunctional materials. Based on a three-dimensional (3D) printer, the new device can produce three-dimensional structures that emulate biological tissue and can change their geometry or properties when stimulated by magnetic fields or electric currents. The researchers also created a self-healing material that "consists of a soft polymer matrix embedded with magnetic particles with a remanent field," according to UC3M's Daniel García González. "For practical purposes, it is as if we had small magnets distributed in the material, so that if it breaks, when the resulting parts are brought together again, they will physically join, recovering their structural integrity."
|
Top Programming Languages, Topics: Here's What Developers Want to Learn
ZDNet Liam Tung March 3, 2023
O'Reilly Media's Technology Trends for 2023 report, in analyzing data on its learning platform's 2.8 million users, found artificial intelligence was the most frequent topic of interest among developers last year. The learning platform said its users expressed a 42% year-over-year increase in interest in content about natural language processing, and a 23% increase in interest in content about deep learning. However, interest in content about reinforcement learning and chatbots was down 14% and 5.8%, respectively. As for the popularity of programming languages, Java and Python were at the top of the rankings, followed by Go, C++, JavaScript, C#, C, Rust, Microsoft's JavaScript superset TypeScript, R, Kotlin, and Scala.
|
Breakthrough Enables Perfectly Secure Digital Communications
University of Oxford Department of Engineering Science (U.K.) March 7, 2023
An algorithm developed by researchers at the U.K.'s University of Oxford and Carnegie Mellon University hides sensitive information in a way that makes it impossible to detect anything has been concealed. Based on steganography, the algorithm uses minimum entropy coupling, in which two distributions of data are combined to maximize their mutual information while preserving the individual distributions. Oxford's Christian Schroeder de Witt said, "Our method can be applied to any software that automatically generates content, for instance probabilistic video filters, or meme generators. This could be very valuable, for instance, for journalists and aid workers in countries where the act of encryption is illegal."
|
Astrobiologists Train AI to Find Life on Mars
Nature Amanda Heidt March 6, 2023
An international team of astrobiologists has trained an artificial intelligence (AI) model to search for life on Mars by mapping biosignatures in Chile's Atacama Desert. Starting in 2016, the researchers searched the desert for photosynthetic organisms called endoliths, collecting data like drone footage and DNA sequences to emulate the information that satellites, rovers, and drones are gathering on Mars. The team fed the data into a convolutional neural network and a machine learning algorithm that forecast the likeliest locations for life in the Atacama. This reduced the search area by up to 97% and boosted the probability of discovering life by up to 88%.
|
Phone-Based Measurements Provide Information About the Health of Forests
University of Cambridge (U.K.) March 7, 2023
An algorithm developed by researchers at the U.K.'s University of Cambridge can measure tree diameter accurately and nearly five times faster than manual methods using low-cost, low-resolution smartphone LiDAR sensors. Measurements of tree diameter at chest height are used to assess forest health and carbon sequestration levels. The algorithm, incorporated into a custom-built Android app, can estimate trunk diameter from a single image, generating results in near-real time. The algorithm was trained using image processing and computer vision techniques to distinguish trunks from large branches and determine the direction the trees were leaning, among other things.
|
Wheeled Robot Measures Leaf Angles to Help Breed Better Corn Plants
NC State University News Matt Shipman March 6, 2023
Technology developed by North Carolina State University (NC State) and Iowa State University researchers automates the process of collecting data on leaf angles. This data is important for plant breeders because, according to NC State's Lirong Xiang, "the leaf angle affects how efficient the plant is at performing photosynthesis." AngleNet features a wheel-mounted robotic device that is steered manually down a row of plants to capture multiple stereoscopic images of every plant it passes via four tiers of cameras at different heights. The images are fed into AngleNet software, which calculates the leaf angles at various heights. Said Xiang, "This, in turn, can help [plant breeders] identify genetic lines that have desirable traits – or undesirable traits."
|
System Like 3D GPS Tracks Pill Cams Through GI Tract
New Atlas Paul McClure March 6, 2023
Researchers at Boston’s Brigham and Women's Hospital, the Massachusetts Institute of Technology, and the California Institute of Technology have created a system that three-dimensionally (3D) tracks wireless ingestible devices throughout the gastrointestinal (GI) tract. The ingestible device that maps the GI tract (iMAG) engages with a wireless Bluetooth receiver to correlate field data to spatial location for real-time tracking as it travels through the body. The system produces a 3D magnetic field via electromagnetic coils positioned at the patient's back. The researchers found the iMAG could measure activity in a pig's gut accurately when tracked with magnetic fields and X-rays. Brigham's Giovanni Traverso said, "Such a portable and non-invasive procedure holds the potential for significant clinical benefit without causing patients discomfort."
|
Drones Detect Moss Beds, Changes to Antarctic Climate
Queensland University of Technology (Australia) March 6, 2023
Scientists mapped Antarctic regions with aerial drones last summer to chart the impact of climate change on vegetation. Researchers at Australia's Queensland University of Technology (QUT), University of Wollongong, and New Zealand's Auckland University of Technology are analyzing drone-captured imagery that identified regions with moss and lichen in Antarctic Specially Protected Areas previously overlooked by satellites. The project aims to monitor vegetation via smart sensors and artificial intelligence, simulate microclimates, and generate maps of protected areas and other ice-free locales. QUT's Joanna Burrows said, "Drones could increase the likelihood of successful data collection by allowing remote and fragile ecosystems to be surveyed with low impact and can mitigate some accessibility issues."
|
Detection Stays Ahead of Deepfakes — for Now
IEEE Spectrum Matthew Hutson March 6, 2023
Computer scientists are developing more advanced algorithms for generating synthetic content, at the same time they are creating counter-algorithms to detect such content. Intel's Real-Time Deepfake Detector, slated for release this spring, will include FakeCatcher, which can identify facial changes due to blood flow. Developed by researchers at Intel and Binghamton University, FakeCatcher cannot be reverse-engineered easily to train a generation algorithm to get better at fooling it. Among other detection tools, researchers at the University of Florida developed a system that models the human vocal tract and can determine if an audio recording is biologically plausible. When it comes to detecting synthetic text, the University of Maryland's Tom Goldstein said the diversity in how people use language and a dearth of signal means it likely will lag other forms of detection.
|
Smart Microscopy Works Out Where to Take the Picture
Lund University (Sweden) March 7, 2023
A software method developed by scientists at Sweden's Lund University can pinpoint exactly where to focus a high-resolution microscope to record biological processes. Lund's Pontus Nordenfelt said the smart microscopy technique involves scanning the specimen in low resolution to collect datapoints, then using algorithms to calculate where the microscope should focus to image the desired process. The researchers suggested the method not only saves time, but also ensures things of interest are captured in a manner that cannot be replicated. Explained Lund's Oscar André, "With smart microscopy, you can derive from the data exactly what it is you want to collect, thereby removing the potential for human error from the collection process itself."
|
|