Association for Computing Machinery
Welcome to the June 8, 2016 edition of ACM TechNews, providing timely information for IT professionals three times a week.

Updated versions of the ACM TechNews mobile apps are available for Android phones and tablets (click here) and for iPhones (click here) and iPads (click here).

HEADLINES AT A GLANCE


U.S. Gets Warnings and Advice About the Internet of Things
Computerworld (06/07/16) Patrick Thibodeau

In response to a U.S. Department of Commerce request for comments about the Internet of Things' (IoT) potential, industry groups including ACM's U.S. Public Policy Council and others submitted more than 130 reports detailing positive and negative aspects, which will form the basis of a green paper (a tentative government report). Booz Allen Hamilton stresses the workforce would need to adapt to a reality in which some jobs will become redundant, while demand for others will grow. Trends the consultancy forecasts include a need for data specialists "to analyze and remove noise from data, and privacy officers will need to analyze vulnerabilities and evolve policies." Booz Allen also envisions crowdsourcing becoming more popular as a means for corporations "to access top talent on demand," while emotional intelligence, creativity, "and the ability to deduce meaning from information" will be sought. Meanwhile, the American Bar Association warns of the IoT's potential scale becoming so vast that responding to a "disabling attack" could exceed "the capacity of any application vendor, the largest global device manufacturers, a self-help community within an industrial sector, or even national governments to address." ACM cautions although vendors can secure individual devices, once users start building devices into a "composable" infrastructure, the certainty fades that any safeguards will remain. The Electronic Privacy Information Center notes businesses could gain valuable commercial insights about customers with the IoT.


The Web's Creator Looks to Reinvent It
The New York Times (06/07/16) Quentin Hardy

World Wide Web creator Sir Tim Berners-Lee seeks to rethink the Web in response to the fact it is largely controlled by governments and corporations, and not the democratic tool for content creation and access he originally envisioned. Berners-Lee and other scientists on Tuesday brainstormed the use of newer technologies to establish a more decentralized, private, and reliable Web at the Decentralized Web Summit. They note the Web lacks full decentralization because of how Web pages are created, managed, and named. One advance the researchers discussed with the potential to boost individual control is new payment technologies. For example, if people adapted the ledger system by which digital currencies are employed, a musician might be able to sell records without intermediaries such as iTunes, while news sites could support a system of micropayments for reading a single article, instead of relying on Web ads for revenue. Meanwhile, Internet Archive director Brewster Kahle notes the archive currently saves discontinued Web pages by searching and capturing them and then giving them a new address. He says with a distributed system, "the archive can have all of the versions, because there would be a permanent record located across many sites." Google chief Internet evangelist and former ACM president Vinton G. Cerf says the Web's accountability could be improved using methods digital currencies utilize to record transactions permanently.
View Full Article - May Require Free Registration | Return to Headlines | Share Facebook  LinkedIn  Twitter 


Researchers Demo How to Build Nearly Invisible Backdoor in Computer Chips
Dark Reading (06/06/16) Jai Vijayan

University of Michigan (U-M) researchers have demonstrated a method that would enable a cyberattacker to install a virtually undetectable backdoor on a microprocessor during the fabrication process that could be exploited later to gain access to systems running the compromised chips. The method is the first fabrication-time processor attack of its kind and the first to show an analog attack is substantially smaller and stealthier than a digital attack. The attack involves the addition of a single, altered logic gate to a chip that is ready for fabrication and the use of an extremely stealthy process for triggering changes in the gate's functionality so it eventually acts in a malicious manner. The attack is nearly undetectable because it involves no significant changes to the chip's circuitry or design, according to the U-M researchers. The altered logic gate is designed to switch from off to on when the accumulated charge in the capacitor reaches a certain pre-defined threshold. "If the wire toggles infrequently, the capacitor voltage stays near zero volts due to natural charge leakage," the researchers say. However, when the wire is toggled frequently, the capacitor begins to charge and eventually reaches a voltage threshold that causes the gate to flip to a malicious state.


Microsoft Finds Cancer Clues in Search Queries
The New York Times (06/07/16) John Markoff

Microsoft researchers have shown that by analyzing search engine queries, they may be able to identify pancreatic cancer sufferers before they have been diagnosed. "We asked ourselves, 'If we heard the whispers of people online, would it provide strong evidence or a clue that something's going on?'" says Microsoft's Eric Horvitz. Starting with Bing searches indicating someone had been diagnosed with the disease, the researchers went backward to look for previous inquiries that could have shown the user was experiencing symptoms prior to diagnosis. They estimate they can identify between 5 percent and 15 percent of pancreatic cases, with false positive rates as low as one in 100,000. Horvitz says the basis of the research is the Microsoft team's ability to precisely differentiate between searches that are casual or anxiety-driven, and authentic searches for specific symptoms by people who are undergoing them. One possible next step would be to organize a health service in which users would permit their searches to be gathered, enabling researchers to screen for questions that indicate warning-flag symptoms. "The question [is], 'What might we do? Might there be a Cortana for health some day?'" Horvitz says.
View Full Article - May Require Free Registration | Return to Headlines | Share Facebook  LinkedIn  Twitter 


Advanced Game Theory Goes to Work for Homeland Security
Government Computer News (06/07/16) Patrick Marshall

The University of Southern California's Teamcore Research Group is creating game theory applications to help agencies solve problems related to homeland security. The group is funded by the U.S. Department of Homeland Security and has developed apps for the Transportation Security Administration and the U.S. Coast Guard. In February, the team unveiled Protection Assistant for Wildlife Security (PAWS), which uses artificial intelligence and game theory to catch animal poachers. PAWS can help patrols find the best routes to apprehend poachers by using historical poaching data. Game theory is being similarly applied to protect airport facilities and the surrounding perimeters in a system known as ARMOR. "You want to come up with a randomized method of allocating checkpoints such that the adversary can't quite figure out by doing surveillance exactly where you will be, but at the same time the more important roads will be covered more often," says Teamcore director Milind Tambe. Protect, the Coast Guard's app, is designed to help patrols locate illegal fishing in the Gulf of Mexico, and further development will include data pertaining to weather conditions. Tambe's team currently is working with the U.S. Army and the Coast Guard to create cybersecurity tools using game theory.


Reduce Cyberslacking and Increase Physical Activity With a Tap, a Click, or a Kick
Waterloo News (06/06/16) Nick Manning

University of Waterloo researchers have developed Tap-Kick-Click: Foot Interaction for a Standing Desk, a wearable technology for standing desks that creates a new way of interacting with computers that could reduce cyberslacking and increase healthy movement. The idea behind the project is that computer users at standing desks can increase their physical activity through indirect, discrete two-foot input using a combination of kicks, foot taps, jumps, and standing postures. University of Waterloo professor Daniel Vogel says the movements are tracked using a depth camera and instrumented shoes. He notes the foot-input techniques also could serve as a cyberslacking deterrent by requiring the user to stand in a mildly uncomfortable position while viewing social-networking sites or other distracting content. If the user changes from that position, the distracting content locks and becomes unusable. "Our technique lets people use those sites, but since they need to stand in an uncomfortable pose while viewing them, they're naturally encouraged to keep it brief," Vogel says. The researchers presented the technology this week at the ACM Designing Interactive Systems (DIS) 2016 conference in Brisbane, Australia.


Microsoft Research Comes Up With a Workable Low-End VR System
Network World (06/06/16) Andy Patrizio

Microsoft Research has developed FlashBack, a system that can lower the barrier to entry for virtual reality (VR) systems and make underpowered devices viable VR platforms. FlashBack relies on cached, pre-rendered frames that are displayed based on the user's actions, instead of using real-time frame rendering. Microsoft's researchers say the system provides an eight-fold improved frame rate, 97 times less energy consumption, and a 15-fold latency reduction in mobile devices. In some cases, FlashBack delivers even better frame rates and responsiveness than a tethered head-mounted display configuration on graphically complex scenes, according to Microsoft. The FlashBack system uses "megaframe," in which dynamic objects can be fully pre-rendered, and then it only displays the frames needed relative to the user's position. FlashBack compresses and stores megaframes either in the graphics processing unit's (GPU) video random-access memory, in regular memory, or on the solid-state drive of the devices; frames are decompressed in the GPU's memory when they are needed. The researchers found when using FlashBack, a decoded 4K texture consumes up to 8 MB of memory, but in its compressed state, it needs only 100 KB. The researchers plan to improve the system's ability to handle multiple dynamic objects via caching and compression.


Google's Training Its AI to Be Android's Security Guard
Wired (06/02/16) Cade Metz

Google's Adrian Ludwig, director of Android security, says computer security should manage risk so it can analyze the overall ecosystem and learn to spot potential vulnerabilities on the fly using deep neural networks. Ludwig acknowledges Google does not currently possess a sufficient volume of Android problems to train its neural networks as fully as it would like. Google's Sebastian Porst says his goal is to use a system called Bouncer to completely automate the identification of any vulnerable or malicious apps that might show up on an Android phone. Bouncer analyzes each app uploaded to the Google Play Store, seeking malicious or otherwise buggy code, and then runs each app to analyze behavior. It also plugs into the Google Web crawler to automatically scan Android apps uploaded to random websites, capturing and analyzing any unknown app downloaded to a certain number of phones. Porst says the app-related data gathered by Bouncer is fed into neural networks so the system can discover which combinations of characteristics signal malware. Meanwhile, Google's Jon Larimer and colleagues are constructing a fuzz-testing system that looks for software holes by feeding random inputs, and can concurrently test numerous Android phones. Larimer says his team is investigating neural nets that can identify the structure of each file the system encounters to enable more rigorous testing.


First Experimental Demonstration of a Quantum Enigma Machine
Technology Review (06/04/16)

University of Rochester researcher Daniel Lum and colleagues say they have developed a working "quantum enigma machine" for the first time. The device is capable of sending secure messages using a key that is shorter than the message itself. "We demonstrated the phenomenon with a proof-of-principle experiment to lock 6 bits per photon while using less than 6 bits per photon of secret key," the researchers say. They note the device consists of a photon gun that fires single photons through a type of mask called a spatial light modulator, which superimposes information onto the photon's wavefront. The receiver detects each photon using a light-sensitive array that can pick out the pattern superimposed on the photon, and then subtract the random signal leaving the original message. All of the information encoded on the photon is randomized by a random signal, and because the message is shorter than the key, it also is possible to transmit a new key for encoding the next message. In this manner, the message and the new key are sent simultaneously and both are kept completely secret. The researchers say the technology and techniques developed for quantum key distribution can be applied immediately to building quantum enigma machines, so the technique could be commercialized in the near future.


High-Tech Librarian Knows Its Books
A*STAR Research (06/02/16)

A*STAR researchers say they are developing robot technology designed to relieve librarians of the menial tasks of their job, while enhancing the searching and sorting of books. The latest project from a team in the A*STAR Institute for Infocomm Research involves an autonomous robotic shelf-scanning (AuRoSS) platform. The technology can self-navigate through libraries at night, and scan radio-frequency identification tags to produce reports on missing and out-of-sequence books. The researchers needed to find a way to steer a tall, wheeled robot through complex mazes of library stacks, while keeping a critical distance from shelves at all times. Another key other obstacle involved reading available library maps, considering their resolutions usually are not detailed enough for robot movement. To help track shelves in real time, the researchers assembled a "macro-mini" manipulator, in which the mobile base robot contains an additional small robotic arm. AuRoSS has achieved up to 99-percent scanning accuracy, according to the researchers. "We are improving the robustness and analytics engine and integrating into library operations," says A*STAR's Renjun Li.


When 'Smart' Apps Become Smart for Real
Umea University (Sweden) (06/02/16) Ingrid Soderbergh

Umea University researcher Esteban Guerrero has developed computer-based methods that enable smart applications to recognize and reason about a human's purposeful activities in order to be able to coach them in a meaningful way. Guerrero has taken a starting point in assessment methods used by therapists and developed new generic methods that a computer system can use. The methods are based on activity-theoretical models of human activity and on newly developed argumentation-theoretical frameworks. The methods were implemented in mobile applications that have been tested with older adults for the purpose of evaluating capacity and performance in exercises that aim to measure different aspects of strength and balance. "The methods could be used in...'smart homes,' for example diagnosis and treatment apps that the person can use at home, or an app measuring and evaluating balance and strength for preventing falls in older adults," Guerrero says.


Computing's Search for Quantum Questions
Quanta Magazine (06/02/16) Stephen Ornes

In the wake of Google's 2015 announcement that the D-Wave 2X quantum computer conducted a task 100 million times faster than a classical computer, scientists are plotting out "benchmark problems," or classes of problems that are specifically appropriate to hybrid quantum machines like the D-Wave 2X. A recent Google study proposed scaled-up quantum annealers should outperform classical computers in certain niche disciplines. A quantum annealer is designed to meet the challenge of solving an NP-hard optimization problem, and to arrive at good-enough solutions instead of best solutions. The D-Wave leverages a novel quantum property to "tunnel" from one solution to another in search of the lowest energy state. Swiss Federal Institute of Technology Zurich researcher Matthias Troyer says the Google study makes the most persuasive case thus far that the D-Wave 2X employs quantum effects to tackle problems. However, researchers are skeptical of Google's claim of the D-Wave's vastly superior speed over a classical algorithm. "They looked for problem instances where quantum annealing should work faster than classical annealing, and on those problems where the classical annealer is very slow, the D-Wave worked well," Troyer notes. Texas A&M University's Helmut Katzgraber says now it is up to researchers to identify classical algorithms that could take advantage of the D-Wave 2X's quantum technology.


For Driverless Cars, Citylike Test Sites Offer the Unpredictable
The New York Times (06/04/16) Neal E. Boudette

Automakers are turning to city-like test sites as they race to create the perfect self-driving car. For example, Mcity in Ann Arbor, MI, offers a 32-acre site to recreate Anywhere, USA, complete with simulated city streets, intersections, traffic lights, storefronts, road signs, parking meters, a tunnel, and eventually a railroad crossing. Ford, General Motors, Honda, Toyota, and Nissan are using Mcity as a place for trial and error, outside of the public eye. Mcity and similar sites provide a way to validate the reliability and safety of autonomous cars. "We have to see how these vehicles interact with their surroundings and need repeatable, reliable tests to do that," says Huei Peng, a professor and director of the University of Michigan Mobility Transformation Center. Autonomous vehicles might have to be driven hundreds of millions, or even billions, of miles to reach a statistical certainty of their safety, according to a recent study from the RAND Corporation. The U.S. National Highway Traffic Safety Administration recently allowed several companies to begin test-driving their autonomous vehicles on public roads. "There are real and significant questions about the safety of new technologies," says NHTSA's Mark Rosekind. "The old model of counting vehicle miles and counting crashes and injuries is not sufficient."
View Full Article - May Require Free Registration | Return to Headlines | Share Facebook  LinkedIn  Twitter 


Abstract News © Copyright 2016 INFORMATION, INC.
Powered by Information, Inc.


To submit feedback about ACM TechNews, contact: [email protected]
Current ACM Members: Unsubscribe/Change your email subscription by logging in at myACM.
Non-Members: Unsubscribe