Welcome to the December 29, 2021 edition of ACM TechNews, providing timely information for IT professionals three times a week.
Please note: In observance of the New Year holiday, the next edition of ACM TechNews will be published on Monday, January 3, 2022. 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."
|
|
|
AP Computer Science Principles Diversifying Computer Science
Diverse
Liann Herder
December 27, 2021
The College Board's analysis of 2016 and 2019's high school graduating classes found its Advanced Placement Computer Science Principles (AP CSP) course is boosting computer science (CS) diversity. In 2019, 68% of Black students, 59% of Latinx students, and 60% of first-generation AP CSP enrollees were engaging in an AP science, technology, engineering, or math course for the first time. AP CSP teaches fundamentals of computer technology, the Internet, cybersecurity, and programming languages, plus creative problem-solving. The College Board's Maureen Reyes said AP CSP's only recommended prerequisite course is Algebra I, "to say to students coming from that class, you already have the skills you need." AP CSP learners were three times as likely to become CS majors in college, and twice as likely to take the programming language-focused AP CSA course. Launched in 2,500 schools in 2016, the course is now offered in over 6,000.
|
Queensland Sunflower Crop Planted by Drone
Australian Broadcasting Corp. News (Australia)
Georgie Hewson
December 27, 2021
A drone-planted sunflower crop in the Australian state of Queensland, resulting from 12 experiments, could be the first of its kind. Drone pilot/farmer Roger Woods, whose company deploys agricultural drones for planting and fertilizing crops, said the project faced unique challenges, like spacing sunflowers fairly consistently; the drone dispersed 45,000 seeds per hectare with the goal of 30,000 plants to germinate per hectare. Woods believes drone planting is less impactful to the soil and more economically efficient than conventional farming. Woods said he also hopes the sunflower field will "educate [the public] about new techniques of farming that are much less harsh on the environment than older techniques."
|
Perfecting Pitch Perception
MIT News
Jennifer Michalowski
December 17, 2021
Neuroscientists at the Massachusetts Institute of Technology (MIT) developed a computational model trained using music, voices, and other naturalistic sounds to determine how humans perceive pitch. Their findings could help researchers reproduce pitch perception in cochlear implants. The researchers asked a deep neural network to identify the repetition rate of sounds in a training set to train it to estimate pitch. MIT's Mark Saddler said, "We very nicely replicated many characteristics of human perception ... suggesting that it's using similar cues from the sounds and the cochlear representation to do the task." Among other things, they determined that nerve cells fire in time with the sound vibrations that reach the inner ear. Said MIT's Josh McDermott, "For cochlear implants to produce normal pitch perception, there needs to be a way to reproduce the fine-grained timing information in the auditory nerve."
|
Cincinnati Children's Used VR Heart Surgery to Heal Boy
Cincinnati Enquirer
Terry DeMio
December 27, 2021
A team of Cincinnati Children's Hospital Medical Center (CCHMC) surgeons used virtual reality (VR) to help plan a surgery to repair 12-year-old Brayden Otten's congenital heart defects. The VR system enables surgeons to "walk into" a patient's digital twin organ, visualize its features and complications, and plan repairs in an individualized surgery. CCHMC's Dr. David Morales said VR planning is easier than three-dimensionally-printed organ models, which require reprinting whenever a change is necessary. CCHMC's Dr. Ryan Moore produced a virtual twin of Otten's heart, and Morales was immersed in VR, day after day, planning the procedure. Morales said the VR planning technology has been used for about 15 complex heart surgeries so far.
|
3D Printed Nanomagnets Unveil World of Patterns in the Magnetic Field
University of Cambridge (U.K.)
Vanessa Bismuth
December 16, 2021
An international team led by researchers at the U.K.'s University of Cambridge created magnetic double helices using three-dimensional (3D) printing and microscopy, producing unusual nanoscale topological textures in the magnetic field. The 3D measurements were performed at the PolLux beamline of the Swiss Light Source at the Paul Scherrer Institute in Switzerland. Cambridge's Claire Donnelly said, "This new ability to pattern the magnetic field at this length scale allows us to define what forces will be applied to magnetic materials and to understand how far we can go with patterning these magnetic fields. If we can control those magnetic forces on the nanoscale, we get closer to reaching the same degree of control as we have in two dimensions."
|
Machine Learning Models Quantum Devices
University of Tokyo (Japan)
December 22, 2021
An algorithm developed by researchers at Japan's University of Tokyo can learn the relationship between quantum inputs and outputs to reconstruct the workings of a time-dependent quantum device. The researchers used machine learning and quantum reservoir computing to build the algorithm. Said University of Tokyo's Quoc Hoan Tran, "Many researchers now report that their quantum systems exhibit some kind of memory effect where present states are affected by previous ones. This means that a simple inspection of input and output states cannot describe the time-dependent nature of the system. You could model the system repeatedly after every change in time, but this would be extremely computationally inefficient. Our aim was to embrace this memory effect and use it to our advantage rather than use brute force to overcome it."
|
Method Can Aid Nuclear Explosion Detection
University of Alaska Fairbanks
Rod Boyce
December 17, 2021
The University of Alaska Fairbanks' Alex Witsil has compiled a library of artificial signals to teach computers to detect explosions, including nuclear detonations, using single-microphone infrasound monitors. Current detection algorithms depend on multi-microphone arrays, which are more costly and susceptible to damage. Single infrasound microphones boost detection capability because they are already deployed, and their computers can be trained to identify explosions by using artificial blast signatures like those Witsil created. "The methods we have worked out will allow monitoring agencies to detect explosions from distances of upward of a couple 100 kilometers," Witsil said.
|
Tech Companies Bring Back the Human Touch
The Wall Street Journal
Laura Forman
December 13, 2021
Online technology companies increasingly are adding a human touch to their services. Online dating app The League has hired Ivy League-educated matchmakers, offering two tiers of matchmaking services. Similarly, Match Group has rolled out an option in which users paying $4.99 per week receive two matches handpicked by trained dating coaches from a pool of matches generated by the app's algorithm and based on answers to four dating questions. Meanwhile, online styling service Stitch Fix is allowing customers to select single items from algorithmically personalized shops, and Instagram reportedly will offer a chronological feed that displays content as it is posted by users, as opposed to the current ranking algorithm that prioritizes user preferences and recommended posts. Among food-delivery apps, DoorDash plans to hire employees in New York City to make grocery deliveries in 15 minutes or less, rather than algorithmically assigning work to contract drivers.
*May Require Paid Registration
|
Walk-Through Metal Detectors Can Be Hacked, Research Finds
Gizmodo
Lucas Ropek
December 21, 2021
Researchers at Cisco Talos have identified nine software vulnerabilities in commonly used metal detectors manufactured by Garrett. The vulnerabilities were detected in Garrett's iC module, which provides network connectivity to two popular walk-through detectors. The module is used to control the detectors remotely and perform real-time monitoring and diagnostics. The researchers wrote in a blog post, "An attacker could manipulate this module to remotely monitor statistics on the metal detector, such as whether the alarm has been triggered or how many visitors have walked through. They could also make configuration changes, such as altering the sensitivity level of a device, which potentially poses a security risk to users who rely on these metal detectors." Talos said device users can mitigate the security flaws by updating their iC modules to the latest version of its firmware.
|
Identifying Fake Voice Recordings
Ruhr-Universität Bochum (Germany)
Julia Weiler
December 20, 2021
Joel Frank and Lea Schönherr at Germany's Ruhr-Universität Bochum (RUB) are developing tools to identify artificial intelligence (AI)-generated fake voice recordings. The researchers first compiled a dataset of about 118,000 AI-generated audio deepfakes, comprising roughly 196 hours of English and Japanese content. They then compared the deepfakes with recordings of real speech, and plotted the files as spectrograms showing frequency distribution over time, yielding subtle distinctions in the high frequencies between real and fake files. Frank and Schönherr then programmed algorithms that can distinguish between deepfakes and real speech as a starting point for scientists to devise novel detection methods.
|
This Robotic Inchworm Just Made It to Higher Ground
IEEE Spectrum
Michelle Hampson
December 17, 2021
Researchers at China's Shanghai Jiao Tong University have developed a robot that moves like an inchworm and can transition between horizontal and vertical planes. The robot is equipped with three fiber-reinforced pneumatic actuators to allow precise control of its tail, head, and body. It also features two pressure suckers with a double layer of silicone, which work with the actuators to propel the robot forward. The robot can carry 500 grams on horizontal planes and 20 grams on vertical walls, and reach top speeds of 21 millimeters per second on horizontal planes and 15 millimeters per second on vertical walls. Said Shanghai Jiao Tong University's Guoying Gu, "It is the first time to achieve transition locomotion of a soft mobile robot between horizontal and vertical planes, which may expand the workspace of the soft robot."
|
How Fast Can Quantum Computers Process Information?
The Jerusalem Post (Israel)
December 25, 2021
Physicists at Germany's University of Bonn and the Technion-Israel Institute of Technology have investigated the determinants of quantum-computer information processing speed. The researchers theoretically deduced the minimum time for quantum gates to transform the wave function and the information contained. Technion's Gal Ness said the team "used fast light pulses to create a so-called quantum superposition of two states of [a cesium] atom. Figuratively speaking, the atom behaves as if it had two different colors at the same time." The atom clones were then compared at intervals via quantum interference to ascertain when a significant change of the matter wave transpired. Technion's Yoav Sagi said the results indicated the minimum wave-change time shortens as energy uncertainty increases, and demonstrated a speed limit imposed by the atom's average energy.
|
AI Writes Better Stories When It Works Backwards From an Ending
New Scientist
Matthew Sparkes
December 15, 2021
Researchers at the Georgia Institute of Technology (Georgia Tech) found that when using artificial intelligence (AI) to craft a story, starting at the end can result in more coherent plots. The researchers trained a neural network using thousands of film and television plot summaries, then supplied it with a human-composed ending. The software looks at the last sentence to assess the characters' actions, generates 15 potential preceding sentences to explain their behavior, and ultimately selects one as the most statistically likely or appropriate. This process is repeated as the story is written in reverse. Georgia Tech's Mark Riedl said tests with human readers found these backward-generated stories were 15% more coherent than those written by an AI from the beginning. However, Riedl said, "We’re still not able to do some of the more complicated things you see in the movies and books where you foreshadow really far into the future."
*May Require Paid Registration
|
|
