Welcome to the May 26, 2017 edition of ACM TechNews, providing timely information for IT professionals three times a week.
Please note: In observance of the U.S. Memorial Day holiday, TechNews will not be published on Monday, May 29. Publication will resume Wednesday, May 31. ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).
|
Russian Researchers Claim First Quantum-Safe Blockchain
HPC Wire
Doug Black
May 25, 2017
The Russian Quantum Center on Thursday announced the creation of the first quantum-safe blockchain for securing cryptocurrencies and classified government communications and other sensitive digital transactions. The center says it protects the blockchain via a combination of quantum key distribution (QKD) and post-quantum cryptography, in essence rendering it "un-hackable." The researchers say the technology generates special blocks signed by QKD network-produced quantum keys instead of using traditional digital signatures. Hyperion analyst Steve Conway says the center's claims would demand refined quantum computing capabilities. "The challenges with creating a quantum computer increase dramatically with the number of qubits," Conway notes. He also says a serious deployment of the kinds of blockchain security capabilities the Russian Quantum Center is describing would require using thousands of qubits. "I'd be surprised if this were in the thousands of qubits range, which is what you'd really need for serious cybersecurity," Conway says.
|
Google's AI Program Rattles Chinese Go Master as It Wins Match
The New York Times
Paul Mozur
May 25, 2017
Google DeepMind's AlphaGo computer program has now won two games of Go in a three-game match against champion Chinese player Ke Jie, adding credibility to claims that artificial intelligence (AI) is superior to humans in executing specific, highly complex tasks. Ke says his emotions during gameplay may have contributed to his loss. "After this time, AlphaGo to me is 100-percent perfection, to me AlphaGo is the god of the Go game," Ke says. Scientists have cited AI systems' emotional detachment and singular focus on the tasks at hand as evidence that computers may eventually replace many white-collar employees. However, computer scientists argue human-AI collaboration is a much better and more productive approach. With this in mind, two professional Go players, each assisted by AlphaGo, will compete against each other on Friday. DeepMind co-founder Demis Hassabis says top amateur Go players, with the help of AlphaGo, can generally defeat the program in a match.
|
Waltzing Robot Teaches Beginners How to Dance Like a Pro
New Scientist
Edd Gent
May 24, 2017
Researchers at Tohoku University in Japan have developed a 1.8-meter-tall robotic dance instructor that gently guides novices through routines while adapting to their skill level. The robot is equipped with a force sensor and two laser rangefinders to track the students' movements, which are compared against motion-capture data recorded from professional dancers to judge their performance. As the students progress, the robot gradually reduces the force used to lead them so they become less reliant on its guidance, demonstrating their overall progress to provide encouragement. In tests with volunteers who had never waltzed before, five out of six students showed improvement. The researchers say their work could have implications beyond dancing, from physical rehabilitation to sports training. Imperial College London researcher Etienne Burdet notes the underlying approach already has been adopted in robots used in medical rehabilitation.
|
Low-Income Students Nowhere to Be Found in STEM
U.S. News & World Report
Lauren Camera
May 25, 2017
Andrew Moore, dean of Carnegie Mellon University's (CMU) School of Computer Science, and several college presidents addressed the lack of low-income students in science, technology, engineering, and math (STEM) fields, and how STEM plays into the future workforce in the U.S, at the 2017 U.S. News STEM Solutions conference. "If you look at where we admit students who are going to have the most amazing careers you can imagine, you can pretty much map that against a map of the suburbs of regions of the United States which are rich enough to have strong math and computer science programs," Moore says. He notes CMU is considering how to provide a softer introduction to these fields for students without a robust STEM background because their high schools lacked rigorous math and science courses. Meanwhile, other universities are trying various strategies to encourage STEM enrollment by low-income and other underrepresented groups.
|
NSA Says New Encryption Standards Needed to Resist Quantum Computing
eWeek
Wayne Rash
May 24, 2017
The U.S. National Security Agency (NSA) is developing encryption standards to withstand quantum computing, especially since the industry and markets are heading toward quantum capability, according to the agency's Neal Ziring. He says NSA's concerns are justified by quantum computers' use of quantum superposition, which would enable them to run complex algorithms whose operation is beyond classical digital systems' abilities. Ziring says NSA's priorities include researching quantum-resistant encryption algorithms, and compiling a list of current encryption methods to no longer be employed; the list also includes encryption methods recommended for national security systems. Although commercial users that do not handle classified information do not have to use these stronger encryption methods, they are permitted to do so. The next step for these users is ensuring encryption practices fulfill NSA guidelines by having them comply with the mandates of federal information processing standard 140-2.
|
When Hatred Goes Viral: Inside Social Media's Efforts to Combat Terrorism
Scientific American
Larry Greenemeier
May 24, 2017
Social media companies' success in using refined algorithms to narrow search results and advertising by mining users' words, images, videos, and location data has not been repeated for targeting terrorist incitement and propaganda videos spread online. Although new video "fingerprinting" tools promise to identify extremist videos as soon as they are posted, there remain unanswered questions about how well they can curb their proliferation. One tool under development is eGLYPH software designed by Dartmouth College professor Hany Farid to analyze video, image, and audio files, creating a hash to identify either an entire video or specific scenes within it. The algorithm compares hashes of new video clips posted online against signatures in a database of known terrorism-promoting content, and upon finding a match it can automatically take the content down or flag it for review. However, instead of using eGLYPH, major social media outlets plan to generate their own shared hashed terrorism video databases.
|
Toward Mass-Producible Quantum Computers
MIT News
Larry Hardesty
May 26, 2017
Researchers at the Massachusetts Institute of Technology (MIT), Harvard University, and Sandia National Laboratories have developed a new method for generating targeted defects in complex diamond structures, which could help lead to the development of practical, diamond-based quantum computing devices. The MIT-Harvard team planed a synthetic diamond to a thickness of 200 nanometers, and then etched optical cavities into the diamond's surface to increase the luminosity of the defects' light emissions. The Sandia researchers then shot 20 to 30 silicon ions into each optical cavity, and afterwards the MIT-Harvard Team fired electrons into the diamond to produce more vacancies, and heated the material to induce silicon-vacancy bonding. The researchers say the method has yielded defects within 50 nanometers of their ideal positions, on average. "The dream scenario in quantum information processing is to make an optical circuit to shuttle photonic qubits and then position a quantum memory wherever you need it," says MIT professor Dirk Englund.
|
Top Hacker Conference to Target Voting Machines
Politico
Edward-Isaac Dovere
May 23, 2017
Hacking U.S. voting machines will be a focus at the DEFCON hacker conference in Las Vegas in late July, according to organizers. They say some of the methods that will be tested at the event include remote exploits of network software, disassembling voting-machine hardware to find vulnerabilities, and manipulating devices at polling sites to test combined physical/virtual attacks. DEFCON founder Jeff Moss says it is alarming that no one has demonstrated where voting system weaknesses are, while the spread of misinformation has been abetted by non-disclosure agreements and private contracts with the machines' suppliers. Finnish programmer Harri Hursti says a lack of security standards for U.S. voting machines is another barrier. Former White House Department of Homeland Security (DHS) liaison Jake Braun would like DEFCON's insights to be shared with DHS and Congress. He dismisses voting machine companies' security claims, noting, "Anyone who says they're un-hackable just doesn't know what they're talking about."
|
Google Plans to Demonstrate the Supremacy of Quantum Computing
IEEE Spectrum
Rachel Courtland
May 24, 2017
Google researchers say they plan to boost the volume of superconducting qubits built on integrated circuits (ICs) to create a 7x7 array and push operations to the limits of even the best supercomputers, demonstrating "quantum supremacy" by year's end. The team says it will perform operations on a 49-qubit system that will trigger chaotic evolution yielding what appears to be random output, which classical computers can model for smaller systems. University of California, Santa Barbara professor John Martinis says the qubits constituting the array also could be employed to build larger "universal" quantum systems with error correction, capable of performing useful tasks such as decryption. Martinis says the challenge of scaling up the quantum IC involves maintaining qubits' function without losing fidelity or boosting error rates. "Error rate and scaling tend to kind of compete against each other," he notes. The team also sees the possibility of scaling up systems beyond 50 qubits without error correction.
|
Texas A&M Researchers Tackle Autonomous Vehicle Security
Texas A&M Engineering News
Shraddha Sankhe
May 19, 2017
Researchers at Texas A&M University have developed an intelligent transportation system prototype designed to avoid collisions and prevent cyberattacks on autonomous vehicles. The researchers applied the theory of dynamic watermarking of sensors in autonomous vehicles to prevent malicious attacks. The team says the system includes 10 cameras that record the movement of self-driving prototype vehicles. The vision sensors in the system receive the images and accurately calculate the exact location and orientation of the vehicles. The information then is transmitted to a server, which uses the data to control the vehicles. The researchers also added a random private signal called a "watermark" to the actuators, whose statistical properties were known to every node in the system, while its actual random values were not revealed. If autonomous vehicles are integrated into society, "they will need to be protected against malicious attacks on sensors," says Texas A&M professor P.R. Kumar.
|
RIT Team Creates High-Speed Internet Lane for Emergency Situations
RIT News
Scott Bureau
May 22, 2017
Researchers at the Rochester Institute of Technology are developing Multi Node Label Routing (MNLR), a faster and more reliable way to send and receive large amounts of data through the Internet in order to better respond to emergency situations. The team says the project aims to improve the information flow between emergency responders at the scene of an incident and decision-makers at the office of emergency management. MNLR is designed with an immediate failover mechanism, meaning that if a link or network node fails, it uses an alternate path. The new protocol runs below existing Internet protocols, enabling normal Internet traffic to run undisrupted. The researchers ran data between 27 nodes representing the network of the incident control center, the 911 call center, and the office of emergency management. The Border Gateway Protocol took about 150 seconds to recover from a link failure, while MNLR recovered in less than 30 seconds.
|
The Deaf-Blind Can Now 'Watch' Television Without Intermediaries
Charles III University of Madrid (Spain)
May 18, 2017
Researchers at the Charles III University of Madrid in Spain have developed PervasiveSUB, software that enables people who are both hearing- and sight-impaired to receive and enjoy TV content without intermediaries in real time. The software compiles all of the subtitles of TV channels and sends them to a central server, which forwards them to mobile devices. The data is then transmitted to the Braille line used by deaf-blind people via the GoAll app, which integrates the software. The researchers say the software is compatible with different Braille lines and controls the speed of the subtitles that are captured from the TV broadcast. A group of sight- and hearing-impaired people tested the software with very satisfactory results. The participants highlighted the advantage of being able to access previously unattainable information in real time, and praised the system's ability to process Braille lines and the ability to adjust the reading and viewing speeds.
|
Chris Manning: How Computers Are Learning to Understand Language?
Stanford University
Andrew Myers
May 22, 2017
In an interview, Stanford University professor Christopher Manning says his work on computational linguistics concentrates on the meaning of words. "We create computational models that assign mathematical values to words and groups of words and use them to successfully read text and derive meaning," Manning notes. He says a key challenge has been training natural-language processing algorithms to adapt to new uses of language, such the association of more positive contexts to previously negative word usage, as in the case of the word "terrific." "It picks up on those soft changes in meaning over time," Manning notes. "It learns by examining language as it is used in the world." In terms of machine translation, Manning says researchers leverage a large volume of already translated text as a source of context from which to build probabilistic models for translating new text. He also notes deep-learning approaches have led to major improvements in machine translation quality.
|
|
