Welcome to the October 9, 2019 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."
|
|
|
Using Machine Learning to Hunt Down Cybercriminals
MIT News
Adam Conner-Simons
October 8, 2019
Researchers at the Massachusetts Institute of Technology's Computer Science and Artificial Intelligence Laboratory and the University of California, San Diego have developed a machine learning system to detect hackers of Internet Protocol (IP) addresses. IP hijackers exploit a vulnerability in the Border Gateway Protocol (BGP), through which networks share routing information so that data packets arrive at the right destination. The team extracted data from network operator mailing lists, and historical BGP data from the global routing table, noting malefactor behaviors of malicious actors, then training the model to automatically identify such patterns. The researchers flagged false positives via manual vetting. Akamai Technologies' David Plonka said, "This project could nicely complement the existing best solutions to prevent such abuse that include filtering, antispoofing, coordination via contact databases, and sharing routing policies so that other networks can validate it."
|
Space Station Experiment Makes Meat in Microgravity Using 3D Bioprinter
CNet
Amanda Kooser
October 7, 2019
Israel's Aleph Farms collaborated with Russian laboratory 3D Bioprinting Solutions and two U.S. firms on an International Space Station experiment, in which a three-dimensional (3D) bioprinter fabricated beef steak in space, using cells harvested from live animals. Aleph Farms' Yoav Reisler said the experiment is a proof of concept, demonstrating that food can be produced free of local land and water resources. Reisler added that the company aims to address food waste and supply nutritional resources to people in need. The startup collects cells from live cows, cultivating them in a lab for assembly into muscle tissue viable for human consumption, with the same taste and texture as farmed meat.
|
Computer Kidney Sheds Light on Proper Hydration
Waterloo News
October 7, 2019
Scientists at the University of Waterloo in Canada have developed a computer-modeled kidney to draw insights on how medication affects patients who do not consume enough water. Waterloo's Anita Layton constructed the first computerized model to simulate muscle contractions that shuttle urine from kidney to bladder, revealing that insufficiently hydrated people could injure their kidneys, by simultaneously taking two blood-pressure drugs and an aspirin. The damage occurs when the water balance is insufficient, causing waste buildup that leads to concentrated urine. Said Layton, "We're now a step closer to understanding how water balance is maintained in mammals."
|
Four Models for Smart Cities
Government Computer News
October 7, 2019
Pennsylvania State University researchers have identified four basic smart-city models from cluster analysis of 60 municipal smart-city plans worldwide. The essential services model defines cities through use of mobile networks in emergency management and healthcare. A second model, smart transportation, concerns cities that attempt to control urban congestion through technologies like information technology, communications, public transit, car sharing, and/or autonomous vehicles. Meanwhile, the broad spectrum model is characterized by a high degree of civic participation and an emphasis on management of urban services, like water, sewage, waste, and pollution control. The fourth and most common model is the business ecosystem, which employs technology development to accelerate economic activity, via investment in digital skills training and high-tech businesses.
|
X-Ray Tech Lays Chip Secrets Bare
IEEE Spectrum
Samuel K. Moore
October 7, 2019
Researchers at Switzerland's Swiss Federal Institutes of Technology in Zurich and Lausanne and the University of Southern California (USC) have developed an X-ray imaging technique for exploring the three-dimensional design of a computer chip. The ptychographic X-ray laminography technique could be used by integrated circuit designers to verify that manufactured chips match their designs or by government agencies concerned about “kill switches” or hardware trojans, without the need to damage the chip. USC's Anthony F. J. Levi said, "You can identify the foundry, aspects of the design, who did the design. It's like a fingerprint." The new technique focuses an X-ray beam on a chip polished down to a 20-micrometer thickness and positioned on a scanning stage at a 61-degree tilt; a photon-counting camera captures the diffraction pattern.
|
An AI Pioneer Wants His Algorithms to Understand the 'Why'
Wired
Will Knight
October 8, 2019
Yoshua Bengio, a researcher at the University of Montreal in Canada who is co-recipient of the 2018 ACM A.M. Turing Award for contributions to the development of deep learning, thinks artificial intelligence will not realize its full potential until it can move beyond pattern recognition and learn more about cause and effect, which would make existing AI systems smarter and more efficient. A robot that understands dropping things causes them to break, for example, would not need to toss dozens of vases onto the floor to see what happens to them. Bengio is developing a version of deep learning that can recognize simple cause-and-effect relationships. His team used a dataset that maps causal relationships between real-world phenomena in terms of probabilities. The resulting algorithm essentially forms a hypothesis about which variables are causally related, and then tests how changes to different variables fit the theory.
|
Supercomputer Simulates 77,000 Neurons in the Brain in Real-Time
New Scientist
Edd Gent
October 8, 2019
The SpiNNaker supercomputer at the University of Manchester in the U.K. can now simulate part of the sensory cortex in real-time, using thousands of virtual neurons. This breakthrough is the first time such a complex simulation has run this fast, and it could be an important step towards creating better brains for robots. SpiNNaker is equipped with 57,000 specialized chips with a total of 1 million processing units, known as cores. The system is designed to run programs that simulate how biological neurons behave and interact. "You can really transfer principles and algorithms that you've uncovered in nature into this artificial brain living in a robot," said Markus Diesmann, a researcher at the Jülich Research Center in Germany, who designed the model used by SpiNNaker.
|
New Tool for Developers Can Help Preserve App Users' Privacy
Carnegie Mellon University CyLab Security and Privacy Institute
Daniel Tkacik
October 4, 2019
Researchers at Carnegie Mellon University's CyLab Security and Privacy Institute have created a tool for software developers to make them more mindful of user privacy when designing apps. Their integrated development environment plug-in, or Coconut, features heuristics that automatically identify requests for user data, prompting a popup reminder to the developer to code an annotation, explaining the underlying reasons for the request. Developers can opt to select reasons from a list of pre-written annotations, and a PrivacyChecker compiles all the data practices programmed into the app, coupled with the annotations. Tests with 18 Android developers revealed that Coconut-designed apps better addressed privacy concerns, and developers better understood the apps' data practices.
|
La Trobe University Swaps Textbooks for VR, AR
ZDNet
Aimee Chanthadavong
October 8, 2019
La Trobe University in Australia has announced it will switch out textbooks for augmented reality (AR) and virtual reality (VR) for teaching anatomy as part of a 12-week pilot. Second-year learners studying health, sciences, and biomedicine will use AR to access three-dimensional anatomy images via smartphone, tablet, or computer, and superimpose anatomical structures over a colleague, who can execute movements in parallel with the app, to understand muscular function. La Trobe also said students can use HTC VR headsets to visualize and manipulate anatomical structures. The pilot's objective is to help students enhance spatial awareness, explorative learning, and accessibility to anatomy images, in addition to corresponding text, clinical cases, and quizzes. La Trobe's Aaron McDonald said, "It allows us to re-write group activity for anatomy ... we've designed the activities so students can work in groups with AR over iPads."
|
Soft Robot Programmed to Move Like an Inchworm
University of Toronto Engineering News
Liz Do
October 4, 2019
Researchers at University of Toronto's Faculty of Applied Science & Engineering (U of T Engineering) in Canada have developed a miniature robot that crawls like an inchworm, using electrothermal actuators (ETAs) programmed to respond to fluctuations in electricity or heat. U of T Engineering's Hani Naguib and colleagues are designing soft, adaptable robots that crawl and curl, as potential substitutes for heavy, metallic robots used in manufacturing. The researchers engineered an ETA with a three-dimensional resting state, applying a technique that combined heat, stress relaxation, and curing to expand its repertoire of movements and shapes. Naguib said the robot has been trained to execute worm-like locomotion, "but our innovative approach means we could train robots to mimic many movements—like the wings of a butterfly."
|
Arm Joins with GM, Toyota to Find Common Ground on Car Chips
Reuters
Stephen Nellis
October 8, 2019
Chip technology firm Arm Holdings, General Motors, and Toyota, have formed the Autonomous Vehicle Computing Consortium with several other industry suppliers, to develop common computing systems for autonomous vehicles. Current test vehicles for developing self-driving software employ large, power-hungry processors typical of data centers. Chip companies and automakers agree that chips' power consumption and size require drastic reduction of the equipment to be of use to the general public. The consortium will initially create a common computing architecture, enabling automakers to more easily craft software that operates on chips from different vendors.
|
'Personalized DJ' Plays Songs to Match your Mood
Futurity.org
October 3, 2019
University of Texas at Austin researchers found machine learning can be used to create personalized music playlists that adjust to each user's changing emotions, much like a real-life disc jockey can. UT Austin's Maytal Saar-Tsechansky and Peter Stone engineered the algorithm to execute a sequence of feedback loops by trying out a song, then applying listener-provided ratings when choosing the next song. When adjusting to users' moods, the algorithm not only considers which songs they will enjoy, but the order of play, without seeming arbitrary. As one song plays, the program produces tens of thousands of possible sequences, testing and predicting which one will satisfy the listener most. This process repeats for every following song on the playlist. The program could be adapted to other kinds of media. Said Saar-Tsechansky, “It can work in any case where you're recommending things to humans, experienced in a sequence."
|
Encryption System Protects Data from Quantum Computers
Scientific American
Sophie Bushwick
October 8, 2019
Researchers at IBM have demonstrated a quantum-proof encryption method called CRYSTALS (short for Cryptographic Suite for Algebraic Lattices) that can successfully encrypt a magnetic-tap storage device. CRYSTALS generates its public and private keys with a family of equations called "lattice problems." The team also developed a CRYSTALS modification that should allow the method to encrypt cloud-based data. The company hopes to use this improvement to render the IBM Cloud quantum-proof by 2020. The IBM researchers also made the system open source, allowing anyone interested in protecting their data to try it.
|
|
