Welcome to the December 30, 2020 edition of ACM TechNews, providing timely information for IT professionals three times a week.

Please note: In observance of the New Year holiday, TechNews will not be published on Friday, January 1. Publication will resume Monday, January 4.

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."
How Machines Are Changing the Way Companies Talk
VentureBeat
Khari Johnson
December 27, 2020


Economics and machine learning (ML) researchers at Columbia and Georgia State universities found artificial intelligence is causing companies to change how they communicate. Natural language processing is being used to parse and process text in the financial documents companies must submit to the U.S. Securities and Exchange Commission (SEC), then ML tools are used to abstract text or rate the language used as positive, neutral, or negative. Signals provided by these tools inform advisers, analysts, and investors' decisions, and machine downloads are associated with faster trading after the posting of an SEC filing. According to a paper written by the researchers, "Anecdotal evidence suggests that executives have become aware that their speech patterns and emotions, evaluated by human or software, impact their assessment by investors and analysts."

Full Article

Security cameras. Facial Recognition Tools in Spotlight in New Jersey False-Arrest Case
The Washington Post
Asa Fitch
December 29, 2020


A Black man in New Jersey is suing local authorities for his wrongful arrest due to a false facial recognition match. A police report said Woodbridge, NJ, police in January 2019 submitted for facial recognition analysis to investigators in New Jersey and New York a driver's license photo they had collected from a suspect, who fled the crime scene. Authorities from two other departments responded that they had a "high profile comparison" to Nijeer Parks, who was arrested and spent more than a week in jail. The criminal case against Parks was dismissed in November 2019. Advocates of facial recognition say it is effective for solving many cases, yet the tool has attracted scrutiny in law enforcement in part because of the potential consequences of false matches. Many facial recognition algorithms have been shown to be less accurate on nonwhite people, and some makers of facial recognition products have pulled back on the technology over racial bias concerns.

Full Article
*May Require Paid Registration
First-Ever Quantum Chess Tournament Won by Amazon Researcher
New Scientist
Leah Crane
December 10, 2020


Amazon researcher Aleksander Kubica won the world's first quantum chess tournament during the virtual Practical Quantum Computing (Q2B) conference. Quantum chess incorporates ideas from quantum mechanics, with pieces able to be placed into a superposition of two locations, for instance, or entangled with one another. The winner must capture the opponent's king and make a robust quantum measurement of its location. California Institute of Technology's Spiros Michalakis said, "It's like you're playing in a multiverse but the different boards [in different universes] are connected to each other." Cantwell noted the ultimate goal of quantum chess is to provide a familiar mechanism for teaching the basics of quantum mechanics.

Full Article

An x-ray of a dog. A Quick Look Under the Skin
Technical University of Munich (Germany)
December 28, 2020


Researchers at the Technical University of Munich (TUM) in Germany have developed self-learning algorithms that can help analyze medical image data. The AIMOS (Artificial Intelligence-based Mouse Organ Segmentation) software is centered on the use of artificial neural networks to recognize patterns and find solutions on their own. The algorithms were trained using images of mice, with the goal of enabling the program to evaluate three-dimensional (3D) full-body scans to display the exact position and shape of specific organs. Said TUM's Bjoern Menze, "The result shows that self-learning algorithms are not only faster at analyzing biological image data than humans, but also more accurate.”

Full Article
3D-Printing Hollow Structures with 'Xolography'
IEEE Spectrum
Charles Q. Choi
December 28, 2020


Researchers at Germany's Brandenburg University of Applied Sciences have developed a three-dimensional (3D) printing technique for generating hollow structures. The xolography method produces such structures at significantly greater speeds and higher resolutions than earlier volumetric techniques, using two types of light to print. A rectangular sheet of ultraviolet light first activates a thin layer of molecules within the printing resin to a latent state; then an image of a slice of the object to be printed is projected onto the sheet with white light, solidifying only the activated resin. Xolography enabled the Brandenburg researchers to 3D-print free-floating objects without support structures, as well as a detailed 3-centimeter-wide bust of a person with precisely defined internal anatomical features.

Full Article
Technology Improves Next-Generation Aqueous Flow Batteries
SciTechDaily
December 27, 2020


Water-based zinc/bromine redox flow batteries (ZBBs) developed by researchers at South Korea's Korea Advanced Institute of Science and Technology (KAIST) are designed to overcome the life expectancy of next-generation water-cell batteries. KAIST's Hee-Tak Kim said of the new technology, "Not only is it cheaper than conventional lithium-ion batteries [LIBs], but it can contribute to the expansion of renewable energy and the safe supply of energy storage systems that can run with more than 80% energy efficiency." The researchers said ZBBs represent the highest output and life expectancy compared to Redox flow batteries, and do not have the risk of ignition and fire seen in lithium-ion batteries.

Full Article

Trains in Indonesia. Prototype App 'Muter' Aims to Help Deaf Train Passengers in Indonesia
Jakarta Post
Ni Nyoman Wira
December 30, 2020


The University of Indonesia's Ajeng Riski Anugrah, Nandhika Prayoga, and Grace Elizabeth Kristiani have developed a prototype application to help alert hearing-impaired train passengers to approaching destinations and other information. The app, called "Muter," features vibration, chat, news, and animation-video BISINDO (Indonesian Sign Language) content, based on the developers' collaboration with the Center of Indonesian Sign Language. Users can customize vibration-based alerts based on the distance to their intended railway stations. Through BISINDO, users receive information about Jakarta's commuter trains and regulations; the system also facilitates communication between hearing-impaired and non-hearing-impaired people by visualizing phrases and words in sign language.

Full Article
*May Require Paid Registration
Tech Platforms Vowed to Address Racial Equity: How Have They Fared?
The Guardian (U.K.)
Kari Paul
December 28, 2020


Facebook and other major technology platforms pledged to improve their diversity and antiracism efforts this year in response to racial unrest in the U.S. Facebook vowed to boost the number of Black people in leadership positions by 30% in the next five years, and to double its number of Black and Latinx employees overall by 2023. Despite hefty commitments to race-positive and diverse organizations and businesses, Facebook is still criticized for permitting hate speech on its platform, and for not strictly enforcing bans on calls to violence. Twitter committed to diverse hires, targeting women to represent half of its workforce by 2025, and for "underrepresented minorities" to represent 25% of its U.S. workforce. As of May, women represented just 42% of its workforce, Latinx 5.1%, and Blacks 6.3%. The company still faces criticism for not addressing significant hate speech on its platform, and for allowing the circulation of misinformation.

Full Article

A U.S. Army aerial drone. U.S. Army Looks to Improve Quadrotor Drone Performance
U.S. Army Research Laboratory
December 8, 2020


The U.S. Army Combat Capabilities Development Command's Army Research Laboratory (ARL) and Rensselaer Polytechnic Institute researchers have developed a trajectory planner to speed the switch between hovering and forward flight in vertical takeoff and landing tail-sitter drones. The planner was engineered for the Army's Common Research Configuration (CRC) platform, a quadrotor biplane tail-sitter used in testing new design features and examining fundamental aerodynamics. The research team thinks the planner may eventually enable the CRC to intelligently switch between hover and forward flight while navigating across dense or urban areas. Said ARL’s Jean-Paul Reddinger, "This method ... is a step in the direction of integrating high-level autonomy with platform-specific dynamics."

Full Article
Extremely Energy Efficient Microprocessor Developed Using Superconductors
EurekAlert
December 28, 2020


Researchers from Japan's Yokohama National University (YNU) have developed a prototype microprocessor using superconductors far more energy-efficient than state-of-the-art microprocessors. The team designed a 4-bit adiabatic quantum-flux-parametron (AQFP) called Monolithic Adiabatic iNtegration Architecture, which YNU's Christopher Ayala called "the world's first adiabatic superconductor microprocessor." He said the AQFP’s data processing operation can run up to a clock frequency of 2.5 gigahertz (GHz), equivalent to that of devices used in modern computing technologies. Even when accounting for the need to cool the chips, Ayala said the AQFP is still about 80 times more energy-efficient than state-of-the-art semiconductor devices in current high-performance computer processors.

Full Article

The battery-free Wiliot Internet of Things tag. Battery-Free IoT: These Tiny Printable Computers Harvest Energy From Radio Waves
Forbes
John Koetsier
December 28, 2020


The Wiliot Internet of Things (IoT) tag is a printable chip with random-access memory, read-only memory, onboard sensors, certified Bluetooth, an ARM central-processing unit, flash memory, and secure communications. The chip, made by fabless semiconductor company Wiliot, is battery-free, harvesting energy from ambient radio waves; it can be glued onto antennas, with input supplied from sensors for temperature or motion or even chemical changes, and output in encrypted Bluetooth-based communications. Without a battery, the device is smaller, more environmentally friendly, and less expensive. The tag is expected to eventually cost just pennies, but Wiliot's Stephen Statler said the real advance is lowering the cost of sensing infrastructure, which is critical to realizing a ubiquitous IoT.

Full Article
January 2021 Issue of Communications of the ACM
 
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]