Welcome to the June 18, 2018 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."
|
|
|
IIT Hyderabad Builds Dataset to Understand Online User-Engagement
The Hindu (India)
Shubashree Desikan
June 16, 2018
Researchers at the Indian Institute of Technology (IIT) Hyderabad have developed a multilabel video-classification dataset for recognizing boredom, confusion, frustration, and engagement. The Dataset for Affective States in E-Environments (DAiSEE) is made up of 9,068 video clips compiled from 112 individuals. For each of the affective states, there are four levels of labels—very low, low, high, and very high—which are provided by observing the viewer's reactions. "The combination of data and annotations related to user engagement sets the platform for DAiSEE as a specialized dataset," says IIT Hyderabad's Vineeth N. Balasubramanian. Traditionally, affective computing research has focused on anger, disgust, fear, happiness, sadness, and surprise. However, Balasubramanian says the affective states studied by DAiSEE are slightly subtler. The researchers used a crowd-voting method to annotate the dataset, picking the best possible answers using a statistical aggregation technique.
|
Apple and Pixar Created a New File Format for Augmented Reality on Your Phone
Popular Science
Stan Horaczek
June 15, 2018
Apple collaborated with Pixar to develop a new file format for augmented reality (AR) on smartphones. The companies say their USDZ format will be natively supported by the upcoming iOS 12 operating system, and it will package all Universal Scene Description elements required by an application to display three-dimensional (3D) objects. These elements will be integrated into a single, uncompressed zip file that apps can refer to in real time to display an object onscreen and in AR. The iOS 12 launch will be accompanied by the rollout of an upgraded iteration of Apple's ARKit platform, which also will natively support USDZ files. Meanwhile, Apple said it will work with Adobe to combine the USDZ format with Creative Cloud apps so content creators can generate 3D objects in apps and on the Web and enable users to engage with them across AR-capable products.
|
Researchers Investigate Potential Threat to Speech Privacy via Smartphone Motion Sensors
UAB News
Tiffany Westry Womack
June 14, 2018
Researchers at the University of Alabama at Birmingham (UAB) have determined that smartphone motion sensors could be exploited to compromise speech privacy, albeit in limited situations. "Unlike with microphones, users do not have to give newly installed applications permission to use them, making these sensors prime tools for malicious activity," says UAB's Nitesh Saxena. He and UAB's Abhishek Anand used stereo speakers, laptop speakers, smartphone speakers, and live human speech set at different volume levels to assess how audio signals affected a smartphone placed on the same surface as the speaker, on a different surface, and through the air. Built-in laptop speakers affected the accelerometer solely when the laptop and the motion sensor were on the same surface, while motion sensors could be impacted by speech signals from a stereo speaker in the same scenario. Neither smartphone speakers nor human speech could trigger a response in the motion sensors via aerial vibrations.
|
Long-Awaited STEM Report Calls for New Research Institutes
Yale Daily News
Marisa Peryer; Amy Xiong
June 14, 2018
Yale University's University Science Strategy Committee has recommended the creation of several new campus-wide research institutes. After meeting almost 60 times over a 16-month period to identify priorities for science investment that could improve Yale’s standing as a research university, the committee released a report whose top recommendations include the creation of a new institute that will integrate data science with mathematical modeling research. The report also calls for the expansion of the Yale Quantum Institute, as well as the creation of new institutes dedicated to studying the inflammatory basis of disease, environmental change and conservation, and neuroscience. In addition, the report suggests five secondary objectives: climate change, the intersections of computer science and other fields, cancer research, precision medicine, and regenerative medicine. The report also offers four recommendations for investments that would support all STEM fields, focusing on STEM graduate education, STEM diversity, state-of-the art instrumentation and engineering facilities, and core facilities and laboratory spaces.
|
AI Drone Learns to Detect Brawls
IEEE Spectrum
Jeremy Hsu
June 14, 2018
Researchers at the University of Cambridge in the U.K., working with colleagues at the Indian Institute of Science, Bangalore and India's National Institute of Technology, Warangal, have used deep learning to develop a drone surveillance system that automatically detects small groups of people fighting each other. The system uses computer vision software that runs in real time to detect violent individuals, says the University of Cambridge's Amarjot Singh. The researchers trained deep learning algorithms to recognize violent actions by identifying body and limb poses in staged video footage of interns mimicking violence. Singh replaced some of the neural network layers at the front-end with fixed parameters, and used supervised learning toward the back-end, exchanging some of the deep learning process with human engineering input. This allowed the resulting ScatterNet Hybrid Deep Learning (SHDL) network to learn more quickly with less data and less available computing power. The researchers are securing permission from Indian officials to test the system at two upcoming music festivals, and Singh is working to incorporate crowd modeling into the deep learning models.
|
Computer Algorithms Can Test the Dodginess of Published Results
The Economist
June 14, 2018
University of Illinois at Urbana-Champaign researchers have developed a method for algorithmically gauging manipulation of published scientific results, given the mean, standard deviation, and number of data points in a result (all three of which are usually stated as part of such a result). The Complete Recovery of Values in Diophantine Systems (CORVIDS) algorithm is designed to deliver a baseline for rating the apparent plausibility of results if it can rebuild a valid dataset or sets. CORVIDS is tasked with finding all possible combinations of numbers that address the linear equations from which the statistics being analyzed are calculated. To more easily detect anomalies, CORVIDS converts the possible datasets into histograms and configures them along three dimensions. Missing values in the dataset at one end of the scale would imply either incorrect reporting of statistics or problems with the underlying data, which could signal biased collection techniques or even fabrication.
|
GoFly Prize Unveils 10 Winning Designs in $2M Contest for Personal Air Vehicles
GeekWire
Alan Boyle
June 14, 2018
The first phase of the GoFly Prize competition for one-person air vehicles has yielded 10 diverse concepts, says GoFly Prize CEO Gwen Lighter. Phase 1 teams were tasked with designing user-friendly vehicles that could make vertical or near-vertical takeoffs and fly 20 miles. Among the submissions that passed muster were the Georgia Institute of Technology's HummingBuzz, a vehicle that has a motorcycle-style fuselage atop an all-electric, ducted coaxial rotor configuration; a five-rotor airbike from a U.K. team; and an electric-motor-powered vehicle from the Netherlands boasting a canard-wing configuration that accommodates a rider in a motorcycle-style orientation. The second phase of the competition requires teams to flesh out their designs and demonstrate safe vehicle operation, either with a human operator or under autonomous control with a mannequin in the driver's seat.
|
UA Encourages Visually Impaired Teens in STEM
UA News (AZ)
Alexis Blue
June 13, 2018
The University of Arizona's (UA) Project-Based Learning Opportunities and Exploration of Mentorship for Students with Visual Impairments in STEM (POEM) is a U.S. National Science Foundation-funded project designed to introduce visually impaired middle and high school students to career possibilities in science, technology, engineering, and math (STEM). Currently, 10 visually impaired students from across Arizona are participating in the program, which launched this month with a weeklong stay at the UA's Mount Lemmon SkyCenter, where the students participated in an adapted version of Sky School, a K-12 science education program. The goal of the 14-month Project POEM is to "come up with ways to motivate our own students—have them be equipped with the knowledge and ability and power to seek opportunities or become a scientist with visual impairment," says UA's Sunggye Hong.
|
Multiple Lasers Could Be Replaced by a Single Microcomb
Chalmers University of Technology
Michael Nystas
June 12, 2018
A single optical device called a microcomb can replace all the high-performance lasers needed to transmit data over fiber-optic networks, according to researchers at the Chalmers University of Technology in Sweden, and Purdue University. A microcomb produces sharp, equidistant frequency lines in a microphotonic chip, and Chalmers' Victor Torres-Company says these frequencies interfere destructively over a short period, forming a wave within the optical cavity resembling a "hole" of light. "The interesting aspect of this waveform is that it yielded a sufficient amount of power per frequency line, which was essential to achieve these high-performance experiments in fiber communication systems," he notes. Torres-Company suspects the unique properties of these "dark" light pulses will facilitate novel applications in fiber-optic communication systems and spectroscopy.
|
Robot Vision Makes Solar Cell Manufacture More Efficient
SINTEF
June 12, 2018
Researchers at Norway's SINTEF independent research organization and the Norwegian University of Science and Technology have developed a robot equipped with a number of optical sensors that can detect details in the materials used to manufacture solar cells beyond the capabilities of the human eye, enabling manufacturers to create much more effective products. The dedicated sensor package helps the robot recognize defects in quartz crucibles without destroying any aspect of the solar cell in the process. The SINTEF researchers created the new sensor system by integrating a selection of established sensors, such as the confocal white light sensor, and developing a method for the sensors to communicate with each other.
|
Forgotten Corner of Europe Brought Back to Life, Thanks to AI
University of Bristol News
June 7, 2018
University of Bristol researchers used artificial intelligence methods to analyze 47,000 multilingual pages from newspapers dating back to 1873, revealing previously unknown political and cultural events which took place in a forgotten corner of the Austrian Empire that is now divided between Italy and Slovenia. The study's goal was to learn whether historical changes could be detected from the collective content of local newspapers from the Princely County of Gorizia and Gradisca between 1873 and 1914. The patterns that emerged revealed collective trends of a population in the years leading up to World War I and the final years of that Empire. "In this paper we have shown that, in the space of a few decades, the town embraced new ways to communicate, such as the cinema and the telephone, along with modes of transportation, like the car, the airplane, the bicycle, and the train," says researcher Nello Cristianini.
|
|
