Association for Computing Machinery
Welcome to the July 15, 2016 edition of ACM TechNews, providing timely information for IT professionals three times a week.

Updated versions of the ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).

HEADLINES AT A GLANCE


Mobile Phone Data Reveals Literacy Rates in Developing Countries
Technology Review (07/13/16)

Telenor Group Research's Pal Sundsoy has developed a machine-learning algorithm that has learned to determine the literacy of individuals via analysis of their mobile phone records with 70-percent accuracy. The starting point was a standard household poll of 76,000 mobile phone users living in a developing country in Asia, which logged each person's mobile phone number and whether or not they could read. Sundsoy then matched this data set with call data records from the mobile phone company, yielding data such as the numbers each person called or texted, the length of these calls, air time purchases, and cell tower locations. From this information, Sundsoy could ascertain where all the individuals were when they made their calls or texts, who they were calling or texting, the number of texts received, at what time of day, and so on. He thus built a social network for each user. Sundsoy employed 75 percent of the data to seek patterns associated with illiterate users, using number-crunching and machine learning techniques; the remaining quarter was used to test the possibility of using these patterns to identify illiterate people and regions where a higher proportion of illiterate people are present. Illiteracy predictors Sundsoy's algorithm has found include the location where people spend most of their time, the number of incoming texts and how they differ from the number of outgoing texts, and the social network.


Apple's New App Will Teach the Next Generation How to Code
Wired (07/14/16) Cade Metz

Apple's new Swift Playgrounds iPad application is designed to teach children and novices how to code, using the Swift programming language with their mobile devices. "Swift is not just a thing that pro developers can use," says Apple's Wiley Hodges. "It could be someone's first programming language." Swift Playgrounds is scheduled to become available in the App Store this fall, and it teaches users basic programming concepts such as commands, functions, loops, algorithms, variables, and types using visual metaphors as well as actual code. The visual metaphors are playgrounds through which users manipulate the movements and actions of a character named "Byte," using commands. The lessons the app brings to the iPad are particularly geared toward children who are growing up with touchscreens instead of keyboards as their standard computing interface. Users also can construct their own playground environments, where they can see the results of any of their own code.


DARPA Challenge Tests AI as Cybersecurity Defenders
IEEE Spectrum (07/13/16) Jeremy Hsu

The U.S. Defense Advanced Research Projects Agency (DARPA) created the Cyber Grand Challenge to determine whether artificial intelligence (AI) can find and correct "zero-day" software flaws much faster than human cybersecurity experts. This summer, seven teams will compete using AI systems that can autonomously scan rivals' network servers for exploits and shield their own servers by finding and fixing software bugs. DARPA hopes the results of the contest will validate AI systems as sufficiently capable of helping humans safeguard computer software and networks. The teams are using high-performance computers to develop AI "cyber-reasoning" systems, and during the finals each competing AI will be tasked with learning an assigned network's software language and logic before they start determining what bugs are present. To launch attacks against rivals, the AIs will first scan for bugs in their network servers, and alert a DARPA referee about any flaws they find and predict when such a software crash might occur. Meanwhile, the AIs will need to create their own network defenses to protect their servers against rivals. To offset the possibility of such systems being misused, the results and competitor software will be open sourced. "We're hoping to see proof that the entire computer security cycle of responding to flaws can be automated," says DARPA program manager Mike Walker.


Driverless Cars Learn From Landscape Pics Before Going Off-Road
New Scientist (07/13/16) Conor Gearin

New York University (NYU) researchers have taught a machine-learning system to drive off-road by showing it photos of different landscapes. The researchers, led by NYU's Artem Provodin, used pictures of different landscapes from the ImageNet database to teach the artificial intelligence system to recognize drivable terrain. They selected a small set of images showing landscapes and labeled them by hand, highlighting areas where a vehicle could be driven and non-navigable areas. The system used these initial examples to learn to label landscape features on its own, and the researchers then refined the system by applying it to a four-wheeled robot called Corobot Jr. The researchers say the robot is not built for off-road driving, and is unable to drive over grass more than a few centimeters tall. Although the images gave Corobot Jr a head start, these types of limitations are best learned through trial and error, according to the researchers. They say when the robot drives over a surface, it adds that to its list of things on which it can drive. The researchers expect the robot will become more independent over time.


Exploring Networks Efficiently
MIT News (07/13/16) Larry Hardesty

Researchers from the Massachusetts Institute of Technology's (MIT) Computer Science and Artificial Intelligence Laboratory will present work at the upcoming ACM Symposium on Principles of Distributed Computing (PODC 2016) conference in Chicago hypothesizing that the study of ant colony behavior could lead to improved network communication algorithms. The research is based on the theory that ants use the frequency with which they bump into other ants while randomly exploring their surroundings as a foundation for population-density estimates. MIT graduate student Cameron Musco and colleagues use a graph of an ant's environment to outline the insect's "random walk," comparing it to random sampling. They found both methods' accuracy gets better with each additional sample while also converging on true population density at similar speeds. Their analytic techniques are applicable to any graph, including one that describes which members of a social network are linked, or which devices in an ad hoc network are within vicinity of each other. Provided two random walks originating from the same node are likely to branch out in different directions, random walks remain virtually as proficient as random sampling, according to the researchers. In addition, an analysis of random walks executed by a single explorer found pooling observations from many explorers would converge on an accurate estimate faster.


Berkeley Lab Scientists Grow Atomically Thin Transistors and Circuits
Berkeley Lab News Center (07/11/16) Dan Krotz

Scientists at the U.S. Department of Energy's Lawrence Berkeley National Laboratory (Berkeley Lab) have developed a way to chemically assemble transistors and circuits that are only a few atoms thick, an advance that could lead to next-generation electronics and computing technologies that are paper-thin. The team controlled the synthesis of a transistor in which narrow channels were etched onto conducting graphene, and a semiconducting material called a transition-metal dichalcogenide was seeded in the blank channels. The researchers say both of these materials are single-layered crystals and atomically thin, so the two-part assembly yielded electronic structures that are essentially two-dimensional (2D). Moreover, the synthesis is able to cover an area a few centimeters long and a few millimeters wide. The team assembled the structure into the logic circuitry of an inverter. The synthesis of the 2D crystals in the wafer scale is compatible with current semiconductor manufacturing, according to the researchers. "This is a big step toward a scalable and repeatable way to build atomically thin electronics or pack more computing power in a smaller area," says Berkeley Lab's Xiang Zhang.


Nanotech 'Tattoo' Can Map Emotions and Monitor Muscle Activity
American Friends of Tel Aviv University (07/11/16)

Tel Aviv University (TAU) researchers have developed a temporary "electronic tattoo" that can measure the activity of muscle and nerve cells. The tattoo is composed of a carbon electrode, an adhesive surface that attaches to the skin, and a nanotechnology-based conductive polymer coating that enhances the electrode's performance. TAU professor Yael Hanein says the device may improve the therapeutic restoration of damaged nerves and tissues, and could lead to new insights about human emotion. Hanein says the new electrode can map emotions by monitoring facial expressions via electric signals received from the facial muscles. "Our tattoo permits patients to carry on with their daily routines, while the electrode monitors their muscle and nerve activity," she says. In addition, the tattoo can be used to monitor the muscle activity of patients with neurodegenerative diseases. "The physiological data measured in specific muscles may be used in the future to indicate the alertness of drivers on the road; patients in rehabilitation following stroke or brain injury may utilize the 'tattoo' to improve muscle control; and amputees may employ it to move artificial limbs with remaining muscles," Hanein says.


Tiny Microchips Enable Extreme Science
NASA News (07/12/16) Lina Tran

Researchers at the U.S. National Aeronautics and Space Administration (NASA) have developed a family of application-specific integrated circuits (ASICs) designed to measure particles in space. The tiny, radiation-resistant microchips are an important part of NASA's Juno spacecraft, which entered Jupiter's orbit on July 4. The microchips aboard Juno are part of the Jupiter Energetic Particle Detector Instrument (JEDI), a state-of-the-art device that will measure the composition of the huge magnetic system surrounding the planet, known as the magnetosphere. The ASICs measure the speed, energy, and position of particles and photons in space with time accuracy down to a fraction of a billionth of a second. Without these chips, satellite electronics would be much heavier and require substantially more shielding and power. The chips will help researchers understand Jupiter's environment and how it was formed and continues to evolve. ASICs have been incorporated in many other NASA missions to study a diverse range of space environments from close to the sun to the heart of Earth's radiation belts to the edge of the solar system. However, the Juno ASICs were specially developed to be radiation-hardened, enabling them to withstand the harsh, radiative environment of Jupiter's magnetosphere.


'Big Data' Study Discovers Earliest Sign of Alzheimer's Development
McGill University (Canada) (07/11/16) Shawn Hayward

A computational analysis of thousands of brain scans has given researchers a greater understanding of the complex interactions between neurological mechanisms in late-onset Alzheimer's disease (LOAD). The study found decreased cerebral blood flow, not an increase in amyloid protein as previously believed, is the first physiological sign of LOAD. Researchers at Montreal Neurological Institute and Hospital have analyzed more than 7,700 brain images from 1,171 people in various stages of LOAD progression, with patient data coming from the Alzheimer's Disease Neuroimaging Initiative. Although previous research was limited in scope and focused on isolated mechanisms, this study factored in amyloid concentration, glucose metabolism, cerebral blood flow, cognition, and brain atrophy in 78 regions. The sample comes from data taken over a 30-year period in order to record the trajectory of each biological factor. Sophisticated software and terabytes of space were needed to compile and analyze the large dataset. "Increasingly, neurology is limited by the ability to take all this information together and make sense of it," says lead researcher Alan Evans. "This creates complex mathematical and statistical challenges but that's where the future of clinical research in the brain lies."


Social Exchange App Might Help Turn Collaboration Into Currency
Penn State News (07/11/16) Matt Swayne

Simple, symmetrical activities organized via smart technology could be the key to timebanking applications, which enable people to negotiate transactions with their time and social exchanges instead of with money. Pennsylvania State University (PSU) researchers have developed a mobile app called WithShare to tap into the potential of coproduction and mutual interest in timebanking. Unlike traditional models of timebanking, in which services rendered earn credits to pay for other services, WithShare emphasizes shared experiences and symmetric activities, so users can propose a joint project or activity that others can join. "The idea really aims at 'in-the-moment' types of activities with low amounts of coordination," says PSU researcher Benjamin V. Hanrahan. "The experience and the benefit to the participants is enhanced if they take part in an activity with someone else." To test the app, about 40 PSU students used WithShare to post planned activities and indicate their interest in other activities. The app's smart technology also tracks users' locations, transactions, and interactions to help them find interesting experiences based on past activities and interest. The app will be rolled out to a larger audience this fall to continue studying the advantages of coproduction over currency exchanges.


Watch Out, Silicon Chips: Molecular Electronics Are Coming
ScienceDaily (07/11/16)

Researchers at the American Technion Society say they have developed a method for growing carbon nanotubes (CNTs) that could lead to molecular electronics that can replace silicon chips. The researchers note CNT technology has been held back by the need to produce them in specific locations on a smooth substrate, in conditions that will lead to the formation of a circuit around them. The Technion researchers have developed a technology that addresses these challenges. The breakthrough also makes it possible to study the dynamic properties of CNTs, including acceleration, resonance, and the transition from softness to hardness. The researchers say the new method could serve as an applicable platform for the integration of nanoelectronics with silicon technologies, and possibly even replace these technologies in molecular electronics. The researchers, led by Technion professor Yuval Yaish, developed a rapid, non-invasive, and scalable technique that enables optical imaging of CNTs. Instead of depending upon CNT chemical properties to bind marker molecules, the new approach relies on the fact the CNT is both a chemical and physical defect on an otherwise flat and uniform surface. "The goal is to integrate CNTs in an integrated circuit of miniaturized electronic components [mainly transistors] on a single chip," Yaish says.


New WSU Virtual Reality Cave Uses Cutting-Edge Technology
Wichita State University (07/11/16)

Wichita State University's (WSU) "the Cave" is a reconfigurable Flex immersive visualization system that researchers and area businesses and manufacturers can use to create mockups of new designs and test data before ever having to build a physical prototype. The Cave is equipped with a fixed-front wall measuring nearly 19 feet wide and 10 feet high, and hinged side walls that can stretch to create an additional 20 feet. The side walls can swing open to display like a flat wall or be moved into a theater or box shape. The Cave is big enough to create an immersive display of something as large as an airplane. "Our primary vision is to transform manufacturing across multiple industries--aerospace and defense, life sciences, and industrial equipment--through innovative technology," says WSU virtual reality lab manager Jeff Fisher. He notes this includes enhancing concept and design processes, user experience testing, virtual prototyping, facilities planning and certification, and significantly shortening product development timelines for various industries. "We are very excited to see what kind of work WSU will do with this cutting-edge technology, which will put them at the forefront of their field," says Mechdyne's Terry Mercer, who helped conceptualize the design. The Cave was created by an alliance between WSU's National Institute for Aviation Research, Dassault Systemes, and Mechdyne.


ASU Researcher Creates System to Control Robots With the Brain
ASU Now (07/08/16) Scott Seckel

An Arizona State University (ASU) professor has developed technology that enables users to control robots with their thoughts. A controller wears a skull cap outfitted with 128 electrodes wired to a computer, and the device records electrical brain activity. For example, if the user thinks about spreading drones out, "we know what part of the brain controls that thought," says Panagiotis Artemiadis, director of the ASU Human-Oriented Robotics and Control Lab. To make them move, Artemiadis says the controller watches on a monitor and thinks and pictures the drones performing various tasks; a wireless system will send the thought to the robot. "We have a motion-capture system that knows where the quads are, and we change their distance, and that's it," Artemiadis says. He notes the system must be calibrated for individual users, and up to four small robots, some of which fly, can be controlled with the brain interface. Artemiadis envisions the technology being used in the future to control drone swarms for complex operations such as search-and-rescue missions.


Abstract News © Copyright 2016 INFORMATION, INC.
Powered by Information, Inc.


To submit feedback about ACM TechNews, contact: [email protected]
Current ACM Members: Unsubscribe/Change your email subscription by logging in at myACM.
Non-Members: Unsubscribe