Welcome to the December 29, 2017 edition of ACM TechNews, providing timely information for IT professionals three times a week.

Please Note: In observance of the upcoming U.S. holiday, TechNews will not be published on Monday, Jan. 1. Publication will resume Wednesday, Jan. 3.

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."

Hand holding up phone with screen reading no signal New App Developed to Locate People in Areas Without a Phone Signal
December 28, 2017

Researchers at the University of Alicante (UA) in Spain have developed a mobile phone application to locate people who have suffered an accident in remote sites lacking a phone signal. The app "can be incorporated to any smartphone and ... without a signal, emits a Wi-Fi signal which in turn acts as a distress beacon over a distance of several kilometers," says UA professor Jose Angel Berna. He notes signal detection is facilitated with a portable receptor featuring an antenna that links to the smartphone of the search party. Following an accident, the victim can activate the app, which transmits the distress signal periodically, providing location coordinates. Berna says the system can enable the optimization of searches and decrease tracking time, which is crucial when taking into account that "in the case of many deceased people, autopsies have revealed that they survived for several hours and did not die instantly, but it had been impossible to locate them on time."

Full Article
NTU Study Finds That Hackers Could Guess Your Phone PIN Using Its Sensor Data
Nanyang Technological University (Singapore)
Lester Kok
December 26, 2017

Researchers at Nanyang Technological University in Singapore have found that combining data collected from six different sensors in smartphones with machine- and deep-learning algorithms exposed the personal identification numbers (PINs) of Android phones with 99.5-percent accuracy within three guesses. The team used the sensors to simulate which number had been pushed by users, based on the phone's tilt and how much light was blocked by the thumb or fingers. They installed a custom application on the phones that culled data from the accelerometer, gyroscope, magnetometer, proximity sensor, barometer, and ambient light sensor. The researchers say their classification algorithm was fed data collected from three individuals, each of whom entered a random set of 70 four-digit PINs on a phone, while simultaneously recording the relevant sensor reactions. The algorithm assigned different weightings of importance to each sensor, depending on how sensitive each was to the different numbers being pressed, which raised the success rate for PIN retrieval.

Full Article
Carving Out Circuits That Click Into Place
King Abdullah University of Science and Technology
December 29, 2017

Researchers at King Abdullah University of Science and Technology (KAUST) in Saudi Arabia say they have developed a modular strategy in which building blocks presenting complementary geometries generate electronic systems via lock-and-key-type construction to make bonding or soldering unnecessary. The team carved various configurations exhibiting different sizes, heights, and angles out of the back of fragile, ultra-thin flexible silicon integrated circuits to form male modules, and then etched grooves, corresponding to inverse replicas of these shapes, into the substrate to produce hosts for the male modules. "This approach makes the entire assembly simple, easy, and highly reliable for conventional and fully flexible electronic systems," says KAUST's Sohail Shaikh. He notes commercially available integrated circuits were converted into unique geometrical shapes that differ based on function, enabling the modules to be physically sorted by either touch or sight. The team currently is exploring the self-assembly of modular components into electronic systems.

Full Article
The Year in Programming Languages
Paul Krill
December 28, 2017

Enhancements to both established and newer programming languages in 2017 were decidedly mixed, with the popularity of Microsoft's TypeScript rising, especially for developers seeking a substitute for JavaScript. Microsoft's launch of its domain-specific Q# quantum computing language also is a significant development, and the language plugs into Visual Studio IDE and is featured in the Quantum Development Kit. Meanwhile, the easier use of C++ was promoted with the publication of C++ 17, whose augmentations include structured bindings and class template argument deduction. Java Development Kit 9 was advanced as the latest deployment of standard Java following the resolution of disputes over its modularity technology. In addition, Kotlin has experienced an upturn this year, partly due to Google's endorsement of the language for building Android applications.

Full Article

News website on a laptop with fake sign, illustrated College Students Come Up With Plug-in to Combat Fake News
Associated Press
Pat Eaton-Robb
December 25, 2017

A team of college students has developed a browser extension designed to alert users of fake and biased news and help guide them to more balanced coverage. The team's software, developed as part of a hackathon hosted at Yale University and designed as a Google Chrome plug-in, displays a warning screen when someone enters a site known to spread fake news, and also notifies a reader if a story shared on social media is false or biased. The extension employs sentiment analysis technology to examine any story that might be posted in a newsfeed, identifying the major players and any political leanings, and then suggests other stories on the same topic that have an alternate viewpoint. The plug-in also gathers browsing data and can display a graph signaling whether users have been reading stories from only one side of a political spectrum. In addition, the extension curates a newsfeed for individual users, showing alternative stories.

Full Article
Quantum Coupling
Tubingen University
Janna Eberhardt
December 21, 2017

Researchers at the University of Tubingen Institute of Physics in Germany say they have successfully coupled magnetically-stored atoms on a chip to a superconducting microwave resonator, which marks a significant step toward the assembly of a hybrid atomic/superconducting quantum system that should facilitate further development of quantum processors and networks. "We can only achieve full functionality via the combination of different quantum building blocks," notes Tubingen professor Jozsef Fortagh. He says in this manner, rapid calculations can be performed using superconducting circuits, although storage is only possible on very short time scales. "We use the functionality and advantages of both components," notes Tubingen's Helge Hattermann. "The combination of the two unequal quantum systems could enable us to create a real quantum processor with superconducting quantum lattices, atomic quantum storage, and photonic [quantum bits]." Fortagh predicts quantum computers and networks of the future will require a hybrid approach and interdisciplinary research and development for full functionality.

Full Article
Novel Algorithm Enables Statistical Analysis of Time Series Data
MIT News
Sara Cody
December 21, 2017

Researchers at the Massachusetts Institute of Technology (MIT) have developed state-space multitaper time-frequency analysis (SS-MT), a unique algorithm they say delivers time series dataset analysis in real time. The team notes SS-MT enables scientists to work in a more informed manner with large, nonstationary datasets so they can not only measure the fluid properties of data but also make formal statistical comparisons between arbitrary data segments. "The algorithm functions similarly to the way a [global-positioning system] calculates your route when driving," says MIT professor Emery Brown. The team tested SS-MT by first analyzing electroencephalogram readings from patients receiving general anesthesia for surgery. The program produced a de-noised spectrogram defining changes in power across frequencies over time, and the researchers also applied SS-MT's inference paradigm to compare different levels of unconsciousness in terms of the differences in the spectral properties of these behavioral states. "The SS-MT analysis produces cleaner, sharper spectrograms," Brown says.

Full Article
DARPA's Spectrum Management Challenge Completes Its First Round
IEEE Spectrum
Michael Koziol
December 22, 2017

The U.S. Defense Advanced Research Projects Agency is investigating the use of machine intelligence to efficiently manage the crowded available radio spectrum under the Spectrum Collaboration Challenge, which recently concluded its initial qualifying round. Nineteen teams out of 30 were found eligible to compete, using Johns Hopkins University's Colosseum facility for the preliminary event. The Colosseum is equipped with 128 clustered and stripped-down radios that broadcast their signals into cables, feeding directly into a 22-server emulator of real-world situations. The facility can model hundreds of scenarios consecutively to determine which spectrum management algorithms work best. The 10 teams with the algorithms that best managed the spectrum each received $750,000 to continue developing their solutions. "We're on a long road to radio autonomy," says Spectrum Collaboration Challenge project manager Paul Tilghman, who notes the funding awarded to first-round winners is intended to serve as seed money to help cultivate the most promising designs.

Full Article

Robotic hands playing piano AI Has Been Creating Music and the Results Are...Weird
PC Magazine
Ben Dickson
December 23, 2017

Projects combining artificial intelligence (AI) with the creative arts are abundant, with new forms of human-machine cooperation seen as potential advances of such research. One example is the Flow Machines initiative funded by the European Research Council, in which algorithms trained with sheet music from hit songs composed an original song of their own. "The challenge in the modern world is to build an AI that is capable of reflecting ... subjectivity," says Amper Music CEO Drew Silverstein. "Interestingly, sometimes, neural networks and purely data-driven approaches are not the right answer." Experts note an AI's interpretation and understanding of art differs significantly from that of humans, in that it learns abstract representations of artistic works with little experiential relevance. Popgun co-founder Jack Nolan envisions the most persuasive use cases for art-generating AIs as those for collaboration with artists. "We think AI will help [artists create art], rather than replace them," he says.

Full Article

Woman sitting alone in a crowd of people Scientists Teach Robots How to Respect Personal Space
Yan Ou
December 21, 2017

Researchers at the National University of San Juan in Argentina are training robots to avoid collisions with humans through impedance control. Using this method, the researchers aimed to regulate the social dynamics between the robot's movements and the interactions of the robot's environment. The researchers first analyzed how a human leader and a human follower interact on a set track with well-defined borders. They then used this information to program a robot to follow a human within the same defined borders, but without impeding on the social forces defined by the human interactions. "Under the hypothesis that moving like a human will be acceptable by humans, it is believed that the proposed control improves the social acceptance of the robot for this kind of interaction," says National University of San Juan professor Daniel Herrera. The researchers think robots are more likely to be accepted if they can be programmed to respect and respond like humans in social interactions.

Full Article
Quantum Computers Barely Exist--Here's Why We're Writing Languages for Them Anyway
Technology Review
Martin Giles
December 22, 2017

The development of high-level quantum computer programming languages is proceeding despite the nascent state of quantum computing. Microsoft's Krysta Svore stresses the need for providing quantum computer coding languages now because the languages written for current, classical computers will be incompatible for quantum systems. Svore says fully leveraging the power of quantum bits will require developers to have quantum languages so they can generate software that takes complete advantage of the quantum computers' capabilities. She notes classical programming languages make it unnecessary for developers to know how computing hardware works, and the current push is to create quantum programming languages that also protect developers from the complexities of quantum hardware. "This will open the door to more people using these machines, which could lead to new fields of inquiry for the quantum research community," predicts the University of Maryland's Xiaodi Wu.

Full Article
*May Require Free Registration

Blue and green speech bubbles Technique to Allow AI to Learn Words in the Flow of Dialogue Developed
Research at Osaka University
December 21, 2017

Researchers at Osaka University in Japan say they have developed a new technique for dialogue systems called lexical acquisition through implicit confirmation, whereby a computer can obtain the category of an unknown word over multiple dialogues by verifying whether or not its predictions are right in the flow of conversation with humans. The team notes the methodology involves the system deciding whether the prediction is correct or not by using the user response following each request, its context, via machine learning. Furthermore, the system's decision performance was found to be enhanced by considering the classification outcomes from dialogues with other users. The researchers say the results of their work point a way toward the creation of dialogue systems in which a computer can become more intelligent via conversation with humans, and which will lead to the development of dialogue systems with the capability of customizing responses to the user's situation.

Full Article
Lensless Camera Captures Highly Detailed 3D Image
R&D Magazine
Kenny Walter
December 21, 2017

Researchers at the University of California, Berkeley have developed the DiffuserCam, a lensless camera that can convert a single two-dimensional image into high-resolution three-dimensional (3D) images. The team says the DiffuserCam prototype, which is a simple diffuser atop an image sensor, could reconstruct 100 million voxels from a 1.3-megapixel image without scanning. "We think the camera could be useful for self-driving cars, where the 3D information can offer a sense of scale, or it could be used with machine-learning algorithms to perform face detection, track people, or automatically classify objects," notes Berkeley professor Laura Waller. She says the DiffuserCam captures the amount of light hitting a pixel on the image sensor, along with the angle from which the light strikes the pixel. The team is working to remove the need to obtain a few images of a moving point of light to calibrate the software before imaging by using the raw data for calibration.

Full Article
January 2018 Issue of Communications of the ACM
ACM Career & Job Center

Association for Computing Machinery

2 Penn Plaza, Suite 701
New York, NY 10121-0701

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]