Welcome to the May 6, 2020 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."
|
|
|
Data Visualization's Breakthrough Moment in the COVID-19 Crisis
Nightingale
Ben Shneiderman
April 30, 2020
The COVID-19 crisis has spurred data visualization researchers and professionals to apply their expertise to combat the pandemic, designing interactive tools to inform the public and guide decision-makers. Johns Hopkins University researchers built a dashboard that displays current worldwide country data, with evolving windows that visualize information via clickable tabs. The New York Times and other publications offer dashboards with animated pandemic information, while Northeastern University's Paul Kahn and Janice Zhang are adding daily updates to an open source database with more than 600 COVID-19 visualizations. Said Steven Drucker of Microsoft Research, "I don't think there's ever been a moment where data, models, and hence visualization has been thrust so much into the center of everyday life."
|
Computer Scientists Create System to Protect Users' Online Data
University of Waterloo Cheriton School of Computer Science
May 1, 2020
Computer scientists Miti Mazmudar and Ian Goldberg at the University of Waterloo's Cheriton School of Computer Science in Canada have invented a software-based system to help ensure data security for Internet users. The Mitigator system features a plugin that users can install in their browsers that will deliver a secure signal when they visit a website verified to process its data in compliance with its privacy policy. Mitigator's decryptor program passes users' data only to privacy policy-compliant programs, enabling users to confirm that the decryptor is the right program. Said Goldberg, "We want the user to get this assurance that the company's software is running correctly and is processing their data properly and not just leaving it lying around on disk to be stolen."
|
Texas A&M Researchers Help Give Robotic Arms Steady Hand for Surgeries
Texas A&M Today
Vandana Suresh
April 29, 2020
Texas A&M University researchers have developed a technique for ensuring precise control of robotic appendages when performing surgery. Texas A&M's Hangue Park and colleagues designed the method to relay distance information from the controller to the robotic fingers that is independent of visual feedback. By passing distinct frequencies of electrical currents onto fingertips via gloves equipped with stimulation probes, the researchers trained users to associate pulse frequency with distance. Tests showed users receiving electrical pulses were more cognizant of the fingers' proximity to underlying surfaces, and could lower their force of contact by about 70% compared to users receiving visual information alone. Park said, "When our technique is ready for use in surgical settings, physicians will be able to intuitively know how far their robotic fingers are from underlying structures."
|
Social Media Faster Than Official Sources to Identify Software Flaws
GCN
Susan Miller
May 4, 2020
A study by computer scientists at the U.S. Department of Energy's Pacific Northwest National Laboratory (PNNL) found software flaws are likely to be discussed on social media prior to their disclosure on government reporting websites, which could potentially jeopardize national security. The researchers estimated 25% of social media discussions of software vulnerabilities from 2015 through 2017 appeared on social-media sites before their citation in the National Vulnerability Database, and it took nearly 90 days on average for the discussed gap to appear in the database. Analysis of the GitHub, Twitter, and Reddit social platforms revealed that GitHub was by far the most likely point where software bug-related discussions originate. Despite the security threat these vulnerabilities present, the PNNL team said they also offer an opportunity for governments to track social-media discourse about software gaps more closely.
|
'This is Certainly Different': Astronauts on Controlling the Dragon Spacecraft Via Touchscreen
TechCrunch
Devin Coldewey
May 4, 2020
The astronauts manning SpaceX’s Crew Dragon capsule to fly to the International Space Station (ISS) later this month will control the capsule using a touchscreen, rather than with traditional dials and buttons. Astronauts Bob Behnken and Doug Hurley have spent countless hours in simulators, and collaborated with SpaceX from early on to design the new spacecraft controls. A prototype Crew Dragon, piloted both autonomously and remotely, has already visited the ISS and returned. The touchscreen design situates all controls and readouts in the same place, and it is easier to implement a software update in the new system than it was previously, when the pots of 20 different knobs on the control panel might need to be rewired manually. Said Hurley, "It was challenging for us and for (SpaceX) at first to work through those different design issues, but we got to a point where the vehicle, from the manual flying standpoint with the touchscreen, flies very well."
|
Coronavirus Tracking Apps Raise Questions About Bluetooth Security
The Wall Street Journal
Catherine Stupp
April 30, 2020
Governments and businesses intend to launch mobile applications that use Bluetooth to track coronavirus infections, but working with Bluetooth raises cybersecurity concerns, despite researchers' assurances that the technology keeps identifying data confidential. European governments recently announced their development of mobile apps to notify individuals who come into close contact with persons with COVID-19, while Singapore and Australia already are using Bluetooth-based tracing apps. Ben Seri at cybersecurity firm Armis said Bluetooth's massive complexity can lead to developer errors. In 2017, he found a flaw—since patched—in how mobile devices handled Bluetooth signals, which hackers could have exploited to move between devices using Bluetooth connections. Eliot Bendinelli at nonprofit Privacy International said mass use of tracing apps could encourage hackers to try to exploit devices with Bluetooth activated in their vicinity to launch remote cyberattacks.
*May Require Paid Registration
|
Yann LeCun, Yoshua Bengio: Self-Supervised Learning is Key to Human-Level Intelligence
VentureBeat
Kyle Wiggers
May 2, 2020
ACM A.M. Turing Award recipients Yann LeCun and Yoshua Bengio say that self-supervised learning could lead to the creation of artificial intelligence (AI) programs that are more humanlike in their reasoning. Speaking at the International Conference on Learning Representation (ICLR) 2020, which took place online, LeCun, Facebook's chief AI scientist, said supervised learning systems will play a diminishing role as self-supervised learning algorithms—those that generate labels from data by exposing relationships between the data's parts, believed to be critical to achieving human-level intelligence—comes into wider use. Meanwhile Bengio, director at the Montreal Institute for Learning Algorithms, predicts new studies will reveal the way high-level semantic variables connect with how the brain processes information, including visual information. Humans communicate these kinds of variables using language, and they could lead to a new generation of deep learning models.
|
Introducing MIDAS: A Baseline for Anomaly Detection in Graphs
KDNuggets
Limarc Ambalina
April 30, 2020
Researchers at the National University of Singapore (NUS) have developed Microcluster-Based Detector of Anomalies in Edge Streams (MIDAS), a novel anomaly detection approach that outperforms baseline protocols. MIDAS can detect anomalies in real time, helping social networks like Twitter and Facebook identify bogus profiles used for spam and phishing, and investigators detect online sexual predators. NUS' Siddharth Bhatia explained that many real-world graphs are dynamic, which means detection techniques founded on static links may overlook the graphs and anomalies' temporal properties. Said Bhatia, "By using a principled hypothesis testing framework, MIDAS provides theoretical bounds on the false positive probability, which earlier methods do not provide." Tests showed MIDAS was 42% to 48% times more accurate, and processed data 162 to 644 times faster, than baseline approaches.
|
Print Your Own Laboratory-Grade Microscope for £15
University of Bath
Vittoria D'Alessio
April 28, 2020
At the University of Bath in the U.K., researchers have developed an open-source design that allows labs around the world to three-dimensionally (3D) print their own precision microscopes. The OpenFlexure Microscope is a fully automated, laboratory-grade device with motorized sample positioning and focus control, and can be constructed for as little as £15 (about $18.50). It is unique among 3D-printed microscopes in its ability to yield high-quality images, according to the university. In addition, it was designed to be easy to use, with an intuitive software interface and simplified alignment procedures. Said Bath’s Richard Bowman, “The medical device industry is very conservatively regulated, and it would be a good thing if all of this new attention (on 3D-printed hardware) means there’s some rethinking done about how we can uphold high safety standards but make it easier to build something if you’re not a mega corporation.”
|
Automating the Search for New 'Curiosity' Algorithms
MIT News
Kim Martineau
April 28, 2020
Massachusetts Institute of Technology (MIT) researchers have used machine learning to identify new algorithms for encoding forms of curiosity. MIT's Ferran Alet said, "We were inspired to use [artificial intelligence] to find algorithms with curiosity strategies that can adapt to a range of environments." The researchers chose a set of basic modules to define their exploration algorithms, selecting about 36 high-level operations to guide the agent to perform tasks like remembering previous inputs, comparing current and past inputs, and using learning methods to change its own modules. The computer integrated up to seven operations at a time to generate computation graphs describing 52,000 algorithms. Two were entirely new, or apparently too obvious or counterintuitive to be of human design, and outperformed human-designed algorithms on a wide range of simulated tasks and environments.
|
Web of Psychological Cues May Tempt People to Reveal More Online
Penn State News
Matt Swayne
April 28, 2020
Pennsylvania State University (PSU) researchers have identified reasons why people who say they are extremely concerned with their online privacy in practice may readily divulge private information online. Certain cues analyzed by the researchers significantly increased the chance that people would reveal private information such as social security numbers or phone numbers. These cues exploit common pre-existing beliefs about authority, bandwagon, reciprocity, sense-of-community, community-building, self-preservation, control, instant gratification, transparency, machine, and mobility. For example, for the authority cue, a graphic signaling that a website is being overseen by a trusted authority may make people more comfortable with turning over private information to the company. Said PSU's Mary Beth Rosson, "People want to do the right thing and they want to protect their privacy, but in the heat of the moment online, they are swayed by these contextual cues."
|
Pocket-Sized Device Tests Blood Samples for Genetic Conditions
New Scientist
Alice Klein
April 22, 2020
Researchers at the Army Medical University in China have developed a smartphone-powered DNA detector that can identify DNA in blood, urine, and other samples, on the spot. The device, powered by heat from the smartphone, takes just 80 minutes to detect DNA based on characteristic mutations or short genetic sequences. Samples loaded into the detector are mixed with pre-filled chemicals that light up or change color if there is any DNA that matches. The signals are detected via the smartphone's camera, and a light box displays the result. The researchers found the device can identify E. coli bacteria in urine, milk, and river water; it also identified a bacterium that attacks kiwi fruit plants in ground-up samples of the plant's leaves. The device was about 97% accurate when compared with standard laboratory methods.
|
Researchers Use Machine Learning to Unearth Underground Instagram 'Pods'
NYU Tandon School of Engineering
April 27, 2020
A team of researchers at New York University (NYU) Tandon School of Engineering and Drexel University has applied machine learning to uncover Instagram "pods," user clusters that manipulate curation algorithms and artificially boost content popularity. The researchers initially analyzed 1.8 million Instagram posts belonging to 111,455 unique accounts, promoted across more than 400 Instagram pods hosted on Twitter's Telegram instant messaging service. The team then trained a machine learning classifier on metadata from pod groups, and Instagram data connected to both the pods and control posts, to detect pod engagement. NYU Tandon's Janith Weerasinghe said, “Already there is evidence of recently increasing adoption of this strategy: the pods we discovered have emerged at an accelerating pace over the last two years.”
|
|
