Welcome to the March 28, 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."

 Charles the robot (right) mimicking the expression of a researcher (left) Meet the Robot That Can Mimic Human Emotion
Cambridge News (UK)
James Laybourn
March 25, 2018


Charles, a robot built by researchers at Cambridge University in the U.K., mimics human facial expressions. Charles is equipped with computer programs and servos linked to a camera that captures the faces of people who interact with it; this footage is analyzed to measure the positions of facial aspects prior to feeding this data to Charles. Charles then uses this information to guide its servos to replicate the actions of human facial muscles. "We've been interested in seeing if we can give computers the ability to understand social signals, to understand facial expressions, tone of voice, body posture, and gesture," says Cambridge professor Peter Robinson. "We thought it would also be interesting to see if the computer system, the machine, could actually exhibit those same characteristics, and see if people engage with it more because it is showing the sort of responses in its facial expressions that a person would show."

Full Article
Smaller and Faster: The Terahertz Computer Chip Is Now Within Reach
ScienceDaily
March 25, 2018


Researchers at the Hebrew University of Jerusalem (HU) in Israel say they have developed a proof-of-concept technology that will enable computers and optical communication devices to operate 100 times faster. "It will now be possible to manufacture any optical device with the precision and cost-effectiveness of flash technology," notes HU's Meir Grajower. The HU team used a Metal-Oxide-Nitride-Oxide-Silicon structure to create a new integrated circuit that employs flash memory technology in microchips, which could possibly enable standard 8- to 16-GHz computers to achieve a 100-fold speed gain and bring the terahertz chip closer to realization. "This discovery could help fill the 'THz gap' and create new and more powerful wireless devices that could transmit data at significantly higher speeds than currently possible," says HU professor Uriel Levy. "In the world of hi-tech advances, this is game-changing technology."

Full Article

3d illustration of a top view on blue cell pattern with red cell nucleus Computers Made From Human Cells Could Tell You When You're Sick
New Scientist
Helen Thomson
March 23, 2018


Researchers at ETH Zurich in Switzerland say they have created the world's most complex biological computer, a group of engineered cells that could be implanted into the human body to detect diseases and deliver treatments. The team notes the new system is made up of nine cells, each containing chemicals that respond to three inputs that are similar to an AND, NOT, and OR system in a traditional electronic circuit. These cells can coordinate their activities by releasing chemicals that pass from one cell to another. When grouped together, the cells configure into a fully programmable circuit that can respond to multiple inputs. The new multicellular approach enables researchers to program the circuit and achieve different computations simply by connecting the nine cells in different configurations, says ETH Zurich professor Martin Fussenegger. "This work addresses one of the most pressing limitations in synthetic biology--a lack of programmable devices," says Angel Moreno at Newcastle University in the U.K.

Full Article
*May Require Paid Subscription

Lifesaver being thrown at a drowning man in the sea Drone Research Project Could Help Swimmers in Difficulty
Irish Times
Barry Roche
March 26, 2018


U-Flyte is a collaboration between researchers at Maynooth University in Ireland and industry partners aimed at assisting the development of drone technology by developing computer systems to overcome current restrictions that limit the distance over which drone operators can control the machines. Drone operators currently are limited to maintaining their drones within a 300-meter circumference, no higher than 120 meters, and within sight at all times. U-Flyte's purpose is to address the current global logjam preventing the wider development of drone operation and the roll-out of commercial services by providing the research, data, and case studies to guide agencies in expanding these limits, says Maynooth University researcher Tim McCarthy. "Experts even foresee drones being used to transport life-saving medical supplies, or coming to the aid of swimmers, making search and rescue operations safer and more efficient than ever before," McCarthy notes. "However, new research is required to ensure that drones can operate safely and securely."

Full Article
Ruby Is Alive and Well and Thinking About the Next 25 Years
TechRadar
Chris Thornett
March 27, 2018


Ruby programming language creator Yukihiro Matsumoto, speaking at the two-day Bath Ruby Conference in Bath, UK, said he is seeking to define the way forward to address performance and scale issues that developers have for not using Ruby. "Faster execution, less code, smaller teams [are] the keys for the [gain in] productivity," he says. Matsumoto notes Ruby 3.0 aims to be three times as fast as Ruby 2.0, and Ruby also is rolling out its Just-In-Time (JIT) technology to enhance performance further. "With MJIT [compiler], certain Ruby [Yet another Ruby Virtual Machine (YARV)] instructions are converted to C code and put into a .c file, which is compiled by GCC or Clang into a .so dynamic library file," said Shannon Skipper, a developer evangelist at Square. "The RubyVM can then use that cached, precompiled native code from the dynamic library the next time the RubyVM sees that same YARV instruction." Matsumoto said he believes scalability should entail generating less code, as "more code is more maintenance, more debugging, more time, less productivity."

Full Article
Women and Minorities in Tech, By the Numbers
Wired
Blanca Myers
March 27, 2018


Programs such as the BRAID Initiative seek to encourage more women and minorities into technology by persuading universities to alter their introductory computer science (CS) courses. However, while the percentage of white CS graduates has fallen slightly since 1992, accompanied by a proportional increase in other groups, population growth cannot explain the persistent lack of diversity in CS majors. White people are earning college degrees at relatively the same rate they were in 1991, while only Hispanics appear to be entering the CS field, even though more blacks and Hispanics are earning college degrees. A 2016 Google study found black and Hispanic students were 1.5 and 1.7 times respectively more likely to have an interest in learning CS, but also were less likely to have access to those resources. Meanwhile, the number of women studying CS has been declining fairly steadily since the 1980s, despite rising demand for CS talent.

Full Article
*May Require Free Registration

Man’s hand holding money to exchange with hacker to unlock computer, illustration Exposed: The Path of Ransomware Payments
ECN Magazine
March 23, 2018


Researchers at New York University, the University of California, San Diego, Princeton University, and Google have provided the first detailed account of the ransomware payment ecosystem, from initial attack to cash-out. The researchers found that South Koreans are disproportionately impacted by ransomware campaigns, with $2.5 million of the $16 million in ransomware payments tracked in the study having been paid in South Korea. The team also learned that most ransomware operators used BTC-E, a Russian bitcoin exchange that has been seized by the U.S. Federal Bureau of Investigation, to convert bitcoin to fiat currencies. The researchers estimate that at least 20,000 individuals have made ransomware payments over the past two years. The team utilized the public nature of bitcoin blockchain technology to trace ransomware payments over two years, and executed real ransomware binaries in a controlled experimental environment, essentially becoming victims themselves and making micropayments to actual ransom wallets to follow the bitcoin trail.

Full Article
New IBM Machine Learning Framework Demonstrates 46x Speedup
Top 500
Michael Feldman
March 26, 2018


IBM researchers have used a new machine learning framework to surpass Google's previous benchmark for training an online advertising application. The IBM team used the new SNAP Machine Learning (SNAP ML) framework and Power9/V100 GPU servers. IBM's application uses an online advertising click log dataset to predict the chances a user will click through to an ad under a given scenario. The dataset holds 4.2-billion training examples, each of which records whether a user has clicked on an online advertisement under various conditions. Google set the previous benchmark by training the application using TensorFlow and CPU-powered cloud servers, achieving a 70-minute turnaround time. IBM attained the same accuracy with just 91.5 seconds of training. IBM says SNAP ML's ability to handle challenging data environments would apply successfully to real-time or near-real-time applications in which the model must be trained on the fly from streaming data.

Full Article
How Twitter Bots Help Fuel Political Feuds
Scientific American
Chris Baraniuk
March 27, 2018


Researchers in the U.S. and China are studying a "misinformation network" related to the 2016 U.S. presidential election, using software to identify links to unverified related claims and spotting 2 million retweets produced by several hundred thousand accounts to spread misinformation during the six months leading up to the election. Another program was used to analyze the timing, text, and other properties of those retweets to determine the probability that a bot was producing them. "As we got closer and closer to the core, we found more and more bots," says Indiana University professor Filippo Menczer. When core accounts referenced fact-checking sites it was usually to discredit them or to falsely state the fact checkers found a bogus claim to be true. Another team is using artificial intelligence to identify social media bots based on certain tweet characteristics, and they say it could correctly recognize bots based on a single tweet with more than 90-percent accuracy.

Full Article
Atomically Thin Light-Emitting Device Opens the Possibility for 'Invisible' Displays
Berkeley News
Brett Israel
March 26, 2018


Engineers at the University of California, Berkeley (UC Berkeley) have built a millimeters-wide bright-light emitting device (LED) that is fully transparent when deactivated. The light-emitting material is a three-atom-thick monolayer semiconductor, and the team laid it on an insulator and placed electrodes on the monolayer and under the insulator to apply an AC signal across the insulator. As the current switches its polarity from positive to negative and back again, both positive and negative charges manifest simultaneously in the semiconductor, generating light. The researchers have demonstrated this mechanism operates in four different monolayer materials, each of which emit different colors of light. "A lot of work remains to be done and a number of challenges need to be overcome to further advance the technology for practical applications," says UC Berkeley professor Ali Javey. "However, this is one step forward by presenting a device architecture for easy injection of both charges into monolayer semiconductors."

Full Article
UCR Researchers Take Up Fight Against Fake News
UCR Today
Sophia Stuart
March 26, 2018


Researchers at the University of California, Riverside (UCR) are developing ways to address problems in social network analysis, including dissemination of malicious misinformation. "Our work aims at the early detection of such articles, especially in cases where we have no external knowledge regarding the validity and veracity of any article," says UCR professor Evangelos E. Papalexakis. The team analyzes social groups in which all interacting members connect by recording, examining, categorizing, and modeling inputs based on tensor decompositions. All multi-aspects are digitally rendered as multidimensional cubes so the system can investigate and "comprehend" what is actually occurring, grading the news' veracity. "The...techniques...capture nuanced patterns that successfully identify different categories of fake news, without using any external knowledge about the validity of any particular article," Papalexakis says. The researchers' work was presented last month at the 11th ACM International Conference on Web Search and Data Mining (WSDM 2018) in Los Angeles, CA.

Full Article
Could These Grain-Sized Computers Using Blockchain Networks Thwart Counterfeiters?
Computerworld
Lucas Mearian
March 20, 2018


IBM researchers have developed granular microcomputers that could work with a blockchain electronic ledger to verify and track any product from its point of origin to merchants all over the world. These cryptographic anchors (crypto-anchors) could be embedded in everyday objects and devices, says Arvind Krishna, head of IBM Research. The crypto-anchors would serve as mobile sensors or transmitters with unique hashes to authenticate a product's origins and contents, ensuring it matches the electronic blockchain record. For liquids or metals that can expand with heat, the crypto-anchors can employ a special optical device and artificial intelligence algorithms to identify an object's optical structure and features. The technology also identifies the presence of DNA sequences in minutes, IBM says. The product-tracking technology will enable new ways of handling food safety, documenting the authenticity of manufactured components, tracking genetically modified products, identifying counterfeit objects, and protecting luxury goods, says Krishna.

Full Article
Pipe-Crawling Robot Will Help Decommission DoE Nuclear Facility
Carnegie Mellon University
Byron Spice
March 20, 2018


Two autonomous robots developed by The Robotics Institute at Carnegie Mellon University (CMU) will soon crawl through pipes at a decommissioned U.S. Department of Energy uranium enrichment plant in Ohio to identify uranium deposits. "This will transform the way measurements of uranium deposits are made from now on," predicts CMU professor William Whittaker. The RadPiper robot is capable of measuring radiation levels more accurately from within the pipe than external techniques, thus reducing hazards to workers. RadPiper uses a "disc-collimated" radiation sensor developed at CMU to count gamma rays, and it will operate initially in pipes measuring between 30 inches and 42 inches in diameter and will characterize radiation levels in each foot-long segment. CMU's Heather Jones says the robot moves through the pipe steadily atop a pair of flexible tracks, and it has LIDAR and a fisheye camera to detect obstructions ahead. After completing a run of pipe, RadPiper automatically returns to its launch point.

Full Article

3d rendering of  bacteria cells Machine Learning Spots Treasure Trove of Elusive Viruses
Nature
Amy Maxmen
March 19, 2018


Researchers have applied artificial intelligence to the discovery of nearly 6,000 previously unknown species of virus. Simon Roux at the U.S. Department of Energy's Joint Genome Institute trained computers to identify viral genetic sequences from the Inoviridae family. He gave a machine-learning algorithm two sets of data, including 805 genomic sequences from known Inoviridae viruses, and about 2,000 sequences from bacteria and other types of virus, so the program could learn to differentiate between them. The model was then fed metagenomic datasets, and the computer retrieved more than 10,000 Inoviridae genomes, clustering them into groups indicative of different species. A separate study conducted at the University of Sao Paulo in Brazil used machine learning to identify viruses in compost piles at the city's zoo, by programming an algorithm to look for a few distinguishing characteristics of virus genomes. Following training, the computer recovered several genomes that appeared to be new.

Full Article
A History of Your ACM
 
ACM Conferences
 

Association for Computing Machinery

2 Penn Plaza, Suite 701
New York, NY 10121-0701
1-800-342-6626
(U.S./Canada)



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]