Welcome to the May 4, 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."
|
|
|
Using Desserts to Decode Computer Science
Monash University (Australia)
May 3, 2022
Scientists at Australia's Monash University have developed a computer science learning framework that uses food to teach fundamental concepts. Logic Bonbon integrates logic operations with edible materials, whose outcome is a liquid-centered dessert. The system features a pre-made hollow bonbon and the option of three different "logic gates" that allow flavored liquids to flow into the bonbon. Said Monash's Jialin Deng, "Through their interactions with the Logic Bonbons, the participants tangibly experience and learn about logic operations and are essentially creating a mini edible computer that requires an input, performs computation, and results in different combinations of outputs while displaying different emoticons and flavors, allowing to the user to experience what computation 'tastes' like."
|
Comparing AI Reasoning with Human Thinking
IEEE Spectrum
Charles Q. Choi
April 27, 2022
Researchers at the Massachusetts Institute of Technology (MIT) and IBM Research have developed a method for comparing the reasoning of artificial intelligence (AI) software with that of human thinking, in order to better understand the AI's decision-making. The Shared Interest technique compares saliency analyses of an AI decision with human-annotated databases. It classifies the AI's reasoning as one of eight patterns, ranging from the AI being completely distracted (making incorrect predictions and not aligning with human reasoning) to making correct predictions and being completely human-aligned. Said MIT's Angie Boggust, "Providing human users with tools to interrogate and understand their machine-learning models is crucial to ensuring machine-learning models can be safely deployed in the real world."
|
Vertical Thinking Broke Bottleneck in Powering High-Performance Computers
Princeton University
May 2, 2022
A team of researchers from Princeton University, Dartmouth College, Intel, and Google have developed a new method of power delivery that meets the needs of modern high-performance computers without sacrificing speed or efficiency. The method works with both small systems and large datacenters. The goal of the research was to deliver power to smaller areas to enable microprocessors to sit closer together, increase efficiency to reduce costs and prevent overheating, and quickly switch power among components to meet the demands of microprocessors. The researchers accomplished these goals by using capacitors rather than magnetics to process power, and by building the systems vertically instead of horizontally. Princeton's Minjie Chen said the result of their efforts is "a fully functioning system 10 times smaller than the best off-the-shelf."
|
Robot 'Chef' Represents Leap Forward in Automated Cooking
Bloomberg
Nina Massey
May 4, 2022
The designers of a robot chef at the U.K.’s University of Cambridge trained it to "taste" foods at different stages of the chewing process to determine their saltiness, which they suggest could help in developing automated or semi-automated food preparation. "We want the robots to understand the concept of taste, which will make them better cooks," said Cambridge’s Arsen Abdulali. People notice changes in taste and texture as they chew, and the robot, which had been trained to make omelets, tasted nine variations of scrambled eggs and tomatoes at three different stages of the chewing process, then produced taste maps of the different meals. The robot used a saltiness sensor attached to a robot arm to mimic the chewing/tasting process, and “tasted” dishes in a grid-like manner.
|
Computer Scientists Help Build a Wooden Dome Made Solely from Waste
ETH Zurich (Switzerland)
Rahel Künzler
April 26, 2022
A team of architects, engineers, and computer scientists at Switzerland's ETH Zurich built a geodesic dome using demolition waste. The team members salvaged plywood panels, wooden beams, steel girders, and plastic piping from an old car depot in Geneva before its demolition. One student programmed an algorithm to calculate the geometry and dimensions of the desired dome based on the available timber, with the goal of utilizing as much of the scrap wood as possible. Another created an online platform for building materials that serves as a digital material passport, with each piece of wood marked with a QR code that links to the platform. Said team leader Catherine De Wolf, “Projects like this one can only succeed if all the stakeholders are on the same page.”
|
High Schooler Invented Low-Cost, Mind-Controlled Prosthetic Arm
Smithsonian
Margaret Osborne
May 2, 2022
Virginia high school senior Benjamin Choi designed a thought-controlled robotic arm, then three-dimensionally (3D) printed the first model in small pieces, and assembled it with bolts and rubber bands. That model responded to brain wave data and head gestures. Choi's latest iteration employs engineering-grade materials and an artificial intelligence (AI)-powered algorithm that interprets a user's brain waves. The prosthesis only costs about $300 to fabricate. It records the brain's electrical activity using forehead- and ear-worn sensors; the former picks up brain wave information that is transmitted to a microchip in the arm, while an AI model interprets the data to predict the user's thoughts.
|
Researchers Develop Smartphone-Powered Microchip for At-Home Medical Diagnostic Testing
University of Minnesota News
May 2, 2022
University of Minnesota (UMN) researchers have developed a smartphone-powered microfluidic chip that can screen liquid samples for viruses and other biomarkers, clearing a path for faster and more affordable at-home diagnostic testing. The researchers placed electrodes close together on the chip to generate electric fields that pull droplets across the chip, producing a liquid "leg" to detect the molecules within; the process requires just one volt of electricity, which enables the chip's activation via near-field communication signals from a smartphone. The researchers are working with Minnesota startup GRIP Molecular Technologies, which manufactures at-home diagnostic devices, to commercialize the microchip platform.
|
Approach May Help Clear Hurdle to Large-Scale Quantum Computing
Harvard Gazette
Juan Siliezar
May 3, 2022
Researchers at Harvard University, the Massachusetts Institute of Technology, QuEra Computing, and Austria's University of Innsbruck have developed an approach for shuffling qubits during the computation process while maintaining their quantum state, resulting in a programmable, error-correcting quantum computer operating at 24 qubits. The process involves an initial pairing of qubits, the creation of a quantum gate that entangles the pairs and stores the information in hyperfine qubits, moving these qubits into new pairs with other atoms in the system to entangle them as well, and repeating the steps to generate different quantum circuits to perform different algorithms. The atoms ultimately become connected in a "cluster state" and act as backups for each other in the event of an error, according to the researchers.
|
Advertising Source May Drive Experience for Some Siri Users
Penn State News
Jonathan F. McVerry
May 2, 2022
A pair of researchers at Pennsylvania State University (Penn State) and The College of New Jersey (TCNJ) found advertisements on voice assistants (VAs) like Siri may be more effective if users perceive the VA as the medium for an ad, rather than its source, although user motives may influence how the ad is perceived. Response to such ads depend on whether the user was informationally or socially motivated, and Penn State's S. Shyam Sundar said socially motivated users view VAs as "companions in their midst." Informationally motivated users tend to react negatively to ads when the VA is the medium compared to a human spokesperson, while TCNJ's Eugene Cho said users are less put off by VA-promoted ads embedded in social exchanges. He suggested negative responses may be rooted in an ad's context and content, rather than the ad itself.
|
Smart Screws Keep Bridges, Machines, Wind Turbines Safe
Fraunhofer-Gesellschaft (Germany)
May 2, 2022
At Germany's Fraunhofer Cluster of Excellence Cognitive Internet Technologies, researchers developed a self-powered Internet of Things (IoT) device that permits remote monitoring of the stability of screw connections in an object or structure. Smart Screw Connection incorporates sensors and radio technology in a device featuring a screw fitted with a washer containing a piezoresistive DiaForce thin film. The sensors record the preload force at three points when the screw is tightened, and changes in the preload force, which alters the electrical resistance in the thin film, are transmitted via radio module on the screw head to a base station. Fraunhofer's Peter Spies said, "No engineer is required to be on site and check each screw individually, as all the data are transmitted via radio to the service station."
|
Method Detects Deepfake Videos with Up to 99% Accuracy
UC Riverside News
Holly Ober
May 3, 2022
University of California, Riverside (UC Riverside) scientists have developed a technique to detect doctored facial expressions in deepfake videos with up to 99% accuracy. The method splits the task along two branches within a deep neural network: the first branch determines facial expressions, and feeds data on facial regions containing the expression to the second branch, an encoder-decoder, which has an Expression Manipulation Detection architecture to detect and localize manipulated regions in the image. UC Riverside’s Ghazal Mazaheri said, “Multi-task learning can leverage prominent features learned by facial expression recognition systems to benefit the training of conventional manipulation detection systems. Such an approach achieves impressive performance in facial expression manipulation detection.”
|
Smart Office Buildings Are Vulnerable to Hacks
The Wall Street Journal
Konrad Putzier
May 3, 2022
Smart office buildings in the U.S. raise concerns about privacy and cybersecurity. Cybersecurity consultants warn that building managers devote little attention to digital security, and the interconnection of smart building systems means accessing a single Internet-connected door can potentially enable hijacking, extortion, or data theft. Lucian Niemeyer at smart-building safety nonprofit Building Cyber Security worries that more criminals will target smart buildings as protections for mobile phones and databases are strengthened. Said Dave Tyson of cybersecurity company Apollo Information Systems Corp., “The bad guys only need to find one way in, and whatever you’ve connected to is now on the table.”
*May Require Paid Registration
|
|
