Welcome to the October 4, 2017 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."

Two smart cars parked on a street Two Intelligent Vehicles Are Better Than One
Swiss Federal Institute of Technology in Lausanne
Anne-Muriel Brouet
October 4, 2017

Researchers at the Swiss Federal Institute of Technology in Lausanne (EPFL), Switzerland say they are working to improve the reliability and fault tolerance of light detection and ranging systems via a combination of the data collected by an intelligent vehicle's cameras, sensors, and navigation and mapping systems with that collected from other intelligent vehicles. The team notes this could extend the view of an intelligent vehicle that is behind another car. The researchers used simulators and road tests to develop a flexible software framework for networking intelligent vehicles so they can interact. EPFL's Milos Vasic developed cooperative perception algorithms that extend an intelligent vehicle's situational awareness by fusing data from onboard sensors with data provided by cooperative vehicles nearby. The researchers used these cooperative framework algorithms for the software framework. Although the researchers only conducted tests on two vehicles, they say their long-term goal is to create a network between multiple vehicles, as well with the roadway infrastructure.

Full Article
Functional Languages Beat Procedural/Object-Oriented
Application Development Trends
David Ramel
October 3, 2017

Researchers at the University of Virginia and the University of California, Davis say they have completed a study evaluating the impact of programming languages on software quality, and found functional languages are superior to procedural/object-oriented languages in some respects. The team analyzed more than 700 GitHub projects containing about 63 million lines of source code. The researchers note empirically assessing software quality is complex, shaped by numerous interacting variables. "By triangulating findings from different methods, and controlling for confounding effects such as team size, project size, and project history, we report that language design does have a significant, but modest effect on software quality," the team says. Among their findings is that disallowing type confusion seems to offer an improvement over permitting it, while static typing is somewhat better than dynamic in functional languages. The team also found functional languages are associated with fewer defects than procedural/scripting languages, and that managed memory usage is better than unmanaged. The research was published in the October edition of Communications of the ACM.

Full Article
Collaborative Software Development Made Easy
Andrew Silver
October 4, 2017

A growing number of scientific researchers are using continuous integration to automate the testing of software code and other laborious tasks. Continuous integration, which is used heavily in the commercial and open source sectors, involves proposed revisions to software code automatically prompting repetitive tasks such as error-checking, sparing developers the time-consuming job of tracking down bugs. Sebastian Neubert at Heidelberg University in Germany says continuous integration guarantees the software pipeline his team uses in their study of subatomic particles remains correct and consistent as they refine its code, which he notes is an "incredibly valuable" protective measure. Bjorn Gruning at the University of Freiburg in Germany says continuous integration services tend to improve code quality, especially on large projects. Meanwhile, a study presented at the 2016 IEEE/ACM International Conference on Automated Software Engineering (ASE 2016) in Singapore estimated about 40 percent of the 34,544 most popular open source projects hosted on GitHub employed continuous integration in some form.

Full Article
Iris May Ease Your Mind as You Step Into a Driverless Car
The Washington Post
Ashley Halsey III
October 3, 2017

With polls showing most riders are concerned about riding in autonomous cars, scientists are investigating how to imbue such systems with trustworthiness. "What we are finding is the more you can embody...human trust interactions into a human-machine experience, the human perception of the safety and security of that machine changes dramatically," says Intel's Jack Weast. A recent study observed how passengers responded to a simulated self-driving car with a soothing, female human voice nicknamed "Iris," and found participants were more likely to anthropomorphize the vehicle with such a feature. "When people attributed a mind to a [car], they tended to think of it as more intelligent, more thoughtful, more capable of doing the things it's intended to do," says study co-author and University of Chicago professor Nicholas Epley. The team also found participants were four times less likely to blame a car's computer with the Iris voice when accidents occurred.

Full Article
*May Require Free Registration

Elementary school students in a computer lab Improving K–12 Computer Science Education
IEEE Spectrum
Prachi Patel
October 2, 2017

Although a shortage of qualified workers is a frequent refrain of the technology industry, Code.org estimates fewer than half of U.S. K-12 schools offer computer science (CS) courses, which a 2016 Google-Gallup survey elevates to 75 percent when accounting for after-school activities and clubs. The latter poll found most students, parents, and educators value CS, but state policies and curricula often fail to prioritize it, much less make CS education mandatory instead of offering it only as an elective. University of Washington professor Ed Lazowska (recipient of the ACM Presidential Award for 2005 and the ACM Distinguished Service Award for 2009) says computational thinking should be cultivated in all citizens, noting, "It is never too early to start learning this." Not only does starting CS basics in kindergarten build a foundation for logic, critical thinking, and creativity, but Code.org's Pat Yongpradit says it also interests girls and minorities early on. Experts say closing a qualified CS educator gap should be the top priority for new funds allocated by the White House.

Full Article
Bristol Scientists Pinpoint the Singularity for Quantum Computers
University of Bristol News
October 3, 2017

Researchers at the University of Bristol in the U.K. say they have determined that reaching the "quantum computational singularity" will take longer than previously believed. The team notes their research applies to the quantum algorithm known as boson sampling, which is designed to be addressed by photons controlled in optical chips. By reworking an old classical algorithm to simulate boson sampling, the Bristol group demonstrated that "we're at a moment in history where it is still possible for classical algorithms to outperform the quantum algorithms that we expect to ultimately be supersonic," says lead Bristol researcher Anthony Laing. Bristol's Alex Neville notes the largest boson sampling experiment disclosed thus far is for five photons, and he says 30 or even 20 photons were previously thought to be sufficient to signal quantum computational supremacy. However, Neville simulated sampling for 20 photons on a laptop, and scaled it up to 30 photons with departmental servers.

Full Article
ASU Engineers to Aid National Efforts to Advance Computing for 2025 and Beyond
Elizabeth Farquhar
September 26, 2017

Researchers at Arizona State University (ASU) are leading two Joint University Microelectronics Program (JUMP) research centers. JUMP, which is a project of the Semiconductor Research Corporation, is a consortium of technology corporations and the U.S. Defense Advanced Research Projects Agency, which is funding five-year exploratory research projects focused on electronic technologies and circuits, subsystems, and multi-scale systems. ASU professors Yu Cao and Jae-sun Seo are part of the Center for Brain-inspired Computing Enabling Autonomous Intelligence, while ASU professor Shimeng Yu is working within the Applications and Systems-Driven Center for Energy-Efficient Integrated NanoTechnologies. The ASU researchers say they will work on building a bridge between biology and electrical engineering, the development of intelligence hardware design, and "in-memory computing," whose goal is to further resolve some of the challenges of distributed intelligence. The team says this work will be combined with new bio-inspired approximation and compression methods to address the power and area constraints of portable systems.

Full Article

Man with laptop over his head USC ISI Researchers Poised to Crack the Code on Reducing Workplace Stress
USC Viterbi News
Caitlin Dawson
October 3, 2017

Researchers led by the Information Sciences Institute (ISI) at the University of Southern California's (USC) Viterbi School of Engineering aim to relieve workplace stress by first understanding its causes using sensors, data, and artificial intelligence. The team will analyze sensor and self-assessment questionnaire information from USC nurses, and then collaborate with behavioral psychologists on interpreting the results. Wearable sensors, cognitive evaluations, and daily polls will be used in the data collection process to track physical and emotional parameters. "Using this data, we can see how your body actually responds to stressors in the environment," says ISI research director Shrikanth Narayanan. The team wants the psychological insights from the data to eventually warrant the deployment of positive workplace interventions that could reduce stress and increase employee well-being and morale in numerous disciplines. "From what we learn, we can think about how to help people cope better, through on-the-spot interventions...or by informing the design of good workplace environments," Narayanan says.

Full Article

Illustration of “skyrmions Fast-Moving Magnetic Particles Could Enable New Form of Data Storage
MIT News
David Chandler
October 2, 2017

Researchers at the Massachusetts Institute of Technology have shown they can create tiny disturbances in magnetic orientation, or skyrmions, at will in specific locations, which they say could be used to make a new, more efficient data storage system. The system focuses on the boundary region between atoms whose magnetic poles are pointing in one direction and those with poles pointing in the other, and this region can move back and forth within the magnetic material. The researchers found the key to creating skyrmions in specific locations is in material defects. They say by introducing a specific kind of defect in the magnetic layer, the skyrmions become pinned to specific locations on the surface. Those surfaces with intentional defects then can be used as a controllable writing surface for data encoded in the skyrmions. The system also could encode data at high speeds, making it a possible substitute for faster memory systems used in random-access memory for computation.

Full Article

Leaves of a tree with fungal infection Phone-Powered AI Spots Sick Plants With Remarkable Accuracy
Matt Simon
October 2, 2017

Researchers at Pennsylvania State University (PSU) say they have designed a smartphone-based neural-network program that can automatically identify diseases in the cassava plant with near-flawless accuracy. The network is based on Google's open source TensorFlow machine-learning library, and Google's Pete Warden notes the TensorFlow Mobile app requires only about 25 million parameters, versus the hundreds of millions some networks need. "It only requires about 11 billion floating point operations to actually calculate its result, and some other networks require hundreds of billions of operations to do a similar job," Warden notes. He also says thanks to transfer learning, the network was trained to recognize cassava leaves on much less data. "It really comes down to the data, because garbage in, garbage out," says PSU's Amanda Ramcharan. She believes the growing affordability of smartphones and the continuing simplification of algorithms will combine to make such tools more widely available to farmers.

Full Article
International Competition Benchmarks Metagenomics Software
University of Maryland
Irene Ying
October 2, 2017

University of Maryland professor Mihai Pop recently helped evaluate the ability of software to organize, classify, and analyze pools of bacterial DNA as a judge in the Critical Assessment of Metagenome Interpretation (CAMI) competition. The contest tasked researchers with reconstructing and analyzing the genomes of a simulated DNA pool resulting from a combination of about 700 microbial genomes and 600 viral genomes with other DNA sources. Pop says the outcomes determined algorithms that assembled a genome using different lengths of smaller DNA fragments did better than those using DNA fragments of a fixed length--but no assemblers did well at picking apart different, yet similar genomes. Classification and profiling also yielded variable software performance. "What we found was that one [software] tool does better in one context, but another does better in another context," Pop notes. "It is important for researchers to know that they need to choose software based on the specific questions they are trying to answer."

Full Article
A Sea of Spinning Electrons
Rutgers Today
Todd Bates
October 1, 2017

Researchers at Rutgers University-New Brunswick and the University of Florida say they have found a sea of electrons that spin in opposing circles, a phenomenon known as the chiral spin mode. "We discovered a new collective spin mode that can be used to transport energy or information with very little energy dissipation, and it can be a platform for building novel electronic devices such as computers and processors," says Rutgers professor Girsh Blumberg. The research involved using a specialized, ultra-sensitive spectrometer to study a prototype three-dimensional topological insulator, while a microscopic theoretical model predicted the chiral spin mode's energy and temperature evolution. The chiral spin mode is borne from the sea of electrons on the insulator's surface, and the electrons' spinning axes are level and perpendicular to their velocity. Chiral spin modes stem naturally from the surface of such materials, but crystalline defects previously made their observation impossible--a problem addressed by the development of ultra-clean crystals.

Full Article
AI Can Predict Lasting Relationships Based on How You Speak to Your Partner
The Conversation
Ian McLoughlin
September 29, 2017

Researchers at the University of Southern California trained a machine-learning algorithm to understand the relationship between vocal characteristics of couples in therapy and the eventual outcome of therapy, and the program predicted those outcomes with more accuracy than human psychologists. "The significance [of this experiment] is revealing how much information about our underlying feelings is encoded in the way we speak--some of it completely unknown to us," writes University of Kent professor Ian McLoughlin in the U.K. He notes although the therapy participants were speaking naturally, the algorithmic analysis uncovered insights into their mutual feelings they were inadvertently "leaking" into their speech. "This may be one of the first steps in using computers to determine what we are really thinking or feeling," McLoughlin says. He notes potential applications include computers counseling humans about potential partners, or detecting leanings toward antisocial behavior and other negative tendencies or psychological conditions.

Full Article
October 2017 Issue of Communications of the ACM
ACM Publications

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: technews@hq.acm.org