Welcome to the April 24, 2009 edition of ACM TechNews, providing timely information for IT professionals three times a week.
HEADLINES AT A GLANCE
Battle of the Brains Crowns World Champions
TG Daily (04/21/09) Kale, Emory
St. Petersburg State University of IT, Mechanics and Optics has won the ACM International Collegiate Programming Contest (ICPC) World Championship for the second straight year. The team of students from St. Petersburg State solved the most problems--nine--correctly in the shortest amount of time. The top 100 university teams participated in the Battle of the Brains in Stockholm, Sweden, using open standard technology to design software to solve real-world problems. "Serious problems call for great minds and solutions demand a can-do spirit," says ICPC executive director Bill Poucher. "They are athletes of innovation." Tsinghua University in China, St. Petersburg State University in Russia, and Saratov State University in Russia took second, third, and fourth places, respectively, and all won Gold medals. The Massachusetts Institute of Technology was the regional champion for North America, finishing in 7th place and winning a Silver medal, and Carnegie Mellon University took 12th place and received a Bronze medal. The ICPC, sponsored by IBM, has grown 800 percent in size since 1997, and participation this year involved 7,109 teams at 1,838 universities from 88 countries on six continents.
Hathaway Advocates for Direct White House Role on Cybersecurity
Computerworld (04/23/09) Vijayan, Jaikumar
President Barack Obama's acting senior director for cyberspace Melissa Hathaway, who recently completed a 60-day review of the U.S. government's cybersecurity readiness, has called for the White House to take a more direct role in coordinating the nation's cybersecurity efforts. Hathaway says cybersecurity needs to be a shared effort between the private and public sector, but the task of leading that effort is "the fundamental responsibility of our government." She says the government's responsibility "transcends" the scope of individual departments and agencies, none of which have a broad enough view to match the wide variety of challenges. "Protecting cyberspace requires strong vision and leadership and will require changes in policy, technology, education, and perhaps law," she says. Hathaway is a former Bush administration aide who has been working as a cybercoordination executive for the U.S. Office of the Director of National Intelligence. Hathaway's review found that the federal government is not "organized appropriately" to address threats in cyberspace, as responsibilities for cyberspace are scattered across too many departments, which have many overlapping missions and authorities. She also stressed the need for better collaboration between the government and the private sector on cybersecurity, since a large portion of the critical cyberinfrastructure is owned by private companies. "The public and private sector's interests are intertwined with a shared responsibility for ensuring a secure, reliable infrastructure upon which businesses and government services depend," she says. The U.S. also needs to find a way of working with other countries to secure cyberspace, Hathaway says.
Simulated Brain Closer to Thought
BBC News (04/22/09) Palmer, Jason
The Blue Brain project is focused on reverse-engineering mammal brains from laboratory data and to develop a computer model down to the level of the molecules that make a brain. The project has successfully created a detailed simulation of a small region of a brain, developed molecule by molecule, based on the results of an experiment performed on real brains. The project's "Blue Brain" has been put in a virtual body, providing the first indications of the molecular and neural basis of thought and memory. The researchers' success in modeling the neocortical column, a portion of the mammalian brain that is responsible for higher brain functions and thoughts, completes the project's first phase. "The thing about the neocortical column is that you can think of it as an isolated processor. It is very much the same from mouse to man--it gets a bit larger [and] a bit wider in humans, but the circuit diagram is very similar," says Blue Brain project leader Henry Markram. He says the simulated neocortical column is being integrated into a virtual reality agent, a simulated animal in a simulated environment, so researchers can observe the detailed activities in the column as the virtual animal explores its surroundings. "It starts to learn things and starts to remember things," Markram says. "We can actually see when it retrieves a memory, and where they retrieved it from because we can trace back every activity of every molecule, every cell, every connection, and see how the memory was formed." The next phase of the project will use a more advanced version of the IBM Blue Gene supercomputer to add all of the molecules and biochemical pathways. Markram presented his research at the recent European Future Technologies Conference in Prague.
Artificial Intelligence Cracks 4,000-Year-Old Mystery
Wired News (04/23/09) Keim, Brandon
University of Washington computer scientists are using artificial intelligence techniques to decipher the ancient Indus script, which only exists as a series of wall carvings made 4,000 years ago in what is now eastern Pakistan and northwest India. Although archaeologists have uncovered about 1,500 unique inscriptions, researchers have been unable to translate the language or understand its purpose. University of Washington researchers led by Rajesh Rao used pattern-analyzing software running a Markov model, a computational tool used to map system dynamics, and fed the program sequences of four spoken languages. The researchers then gave the program samples of four nonspoken communication systems. The program calculated the level of order present in each of the languages and found nonspoken languages were either highly ordered, with symbols and structures following each other in extremely well-defined ways, or were chaotic, while spoken languages fell in the middle. When the Indus script was submitted to the program, it returned with grammatical rules based on patterns of symbol arrangement that were moderately ordered like spoken languages. The program did not return any information on the meaning of the script, but Rao says the analysis provides a foundation for a more comprehensive understanding of Indus script grammar. "The next step is to create a grammar from the data that we have," he says. "Then we can ask, is this grammar similar to those of the Sanskrit or Indo-European or Dravidian languages? This will give us a language to compare it to."
Ultrasound to Go
Technology Review (04/23/09) Gravitz, Lauren
Washington University in St. Louis computer engineers have developed an ultrasound device that can be plugged directly into a smartphone's USB port. The device enables smartphones to capture images and display them directly on the phone's screen. The phone also can be used to send the images to other users. Washington University researchers William Richard and David Zar created the smartphone device from an ultrasound probe they previously developed that can plug into a laptop's USB port. The researchers transferred all of the necessary computing power to the probe and decreased its power consumption to only half a watt. The image the device creates is smaller and of a lower quality than the laptop version, but the researchers believe that it is more than sufficient for many medical applications. The researchers say it could be used in remote rural communities and developing nations, as well as for emergency imaging of patients en route to a hospital. They also say it could be used for battlefield medicine and for home use by patients with chronic illnesses. Zar plans to have open source software available this summer.
National Science Foundation Awards Millions to Fourteen Universities for Cloud Computing Research
National Science Foundation (04/23/09) Cruikshank, Dana W.
The National Science Foundation (NSF) has awarded almost $5 million in grants to 14 universities through its Cluster Exploratory (CLuE) program to use software and services running on a joint IBM/Google computing cloud to investigate innovative research concepts in data-intensive computing. Carnegie Mellon University (CMU) researchers are using cloud computing to delineate the topicality of Web content to more effectively process Web searches. The Google/IBM cluster is being used to crawl the Web and execute the data cleansing and pre-processing required to develop a Web dataset of 1 billion documents to buoy the research. Another CMU project is focused on the development of the Integrated Cluster Computing Architecture for machine translation, which will make up for existing open source toolkits' failure to keep up with the computing infrastructure required for modern "big data" approaches to machine translations. A second machine translation project being funded by the CLuE program is a University of Maryland-College Park effort to convert text from one language into another through the coupling of network analysis with cross-language information retrieval methods. Researchers at the University of California, San Diego are investigating the management and processing of vast spatial data sets on large-scale computer clusters, while the University of California-Santa Barbara's Massive Graphs in Clusters project is developing a software infrastructure that can efficiently answer queries on very large graph datasets. Florida International University (FIU) researchers are exploiting cloud computing to analyze aerial images and objects to help support disaster mitigation and environmental protection. FIU's CLuE initiative will enable students and researchers to precisely code in the TerraFly database in real time.
Rich Musical Pickings With Easier Access to Archives
ICT Results (04/22/09)
The European Union-funded Enabling Access to Sound Archives through Integration, Enrichment, and Retrieval (Easaier) project is developing new methods for accessing sound archives. The project combines system functions within a single user-configurable interface that enables users to access sound archives in a variety of ways. For example, the system can respond to the needs of both amateurs and professionals by providing ways to interact with and retrieve content through a Web-client access point that functions in any Web browser, or through an advanced, standalone user access system. Both applications use semantic metadata, and can provide a variety of information to users, including tempo, key, and technical and background information. Easaier created a music ontology to generate the metadata. In addition to making sound archives more accessible, the project developed tools that enable users to manipulate content in a variety of ways. For example, users can slow down playback without altering pitch and they can separate specific instruments from a piece, which gives them the ability to play a piece in a higher or lower octave to see how the piece is affected. The tools also can be used with speech and multimedia material, and contain functions such as sound-source separation, equalization, noise-reduction algorithms. Easaier also provides methods to synchronize video and audio streams in real time.
NIST Nudges Quantum Computers Toward Commercial Viability
EE Times (04/22/09) Johnson, Colin
Researchers at the National Institute of Standards and Technology (NIST) have demonstrated an error suppression technique that they say makes multistep quantum calculations commercially feasible. "Quantum computers are inherently prone to errors caused by stray electric or magnetic fields, but theorists have shown that if their error rate is low enough--about one in 10,000--fault-tolerant error correction schemes could enable quantum computations of virtually any length," says NIST lead scientist John Bollinger. "We have taken an idea that other people have developed and have experimentally demonstrated that we can use it to suppress the errors in quantum computers enough to meet that one-in-10,000 requirement." The NIST dynamic decoupling technique, called spin-echo error suppression, uses the echo from error-suppression pulses to realign the "spins" inside qubits that simultaneously encode the superimposed ones and zeros. The regular application of error-suppressing pulses ensures that spins edging into error are realigned before they cause a hard error. Bollinger says that NIST is the first to demonstrate spin-echo refinements capable of suppressing qubit errors below the one-in-10,000 level. The spin-echo technique uses feedback on how many errors are being made to change the spacing of the spin-echo pulses for optimal error suppression.
Sporty Software Reveals Where Blurred Balls Are Heading
New Scientist (04/22/09) No. 2704, Battersby, Stephen
Researchers at the Polytechnic of Milan have developed software that is capable of determining where fast-moving balls appearing as blurry streaks in photographs and video stills are headed. The software can detect the angle of the ball in relation to the camera, even if the ball is moving toward or away from the camera, unlike existing software. Alessandro Giusti and his colleagues thought of a blurred image as a series of sharp images added together, decided to calculate what a series of brief exposures would look like, and developed an algorithm that describes the transparent edges of a motion-blurred image. The algorithm can determine the location of the edge of a ball and then calculate the change in its distance from the camera, using information such as the color of the ball and its background. Speed and direction can be worked out from the exposure time and the size of the ball, from an image of relatively short smears, and the spin of the ball can be determined if there is a surface pattern. The software, which would be cheaper than systems that use coordinated video cameras to follow a ball's motion, could be useful as a training tool for golfers or for judging line calls in tennis matches.
Computing Revolution Creating Desktop Supercomputers
ScienceAlert (Australia) (04/22/09) Beidatsch, Karl
University of Western Australia (UWA) Ph.D. candidate Chris Harris is investigating the use of graphics processing units (GPUs) in radio astronomy data processing. Harris says the GPUs typically used in desktop computers could provide the processing backbone for data-intensive applications running on supercomputers. "If you're trying to get [an] image of the sky using an interferometer, you're using multiple telescopes to get a better image," he says. "What that means is that you need to take those separate signals and combine them to form an image, and there's a great deal of computation involved in that." Harris says that GPUs are designed to handle numerous, small processing tasks. "My research shows that when I use a GPU with a parallel correlation algorithm, it's 10 to a hundred times faster," he says. UWA professor Karen Haines says the ability to quickly process data makes GPUs invaluable to projects such as the Square Kilometre Array astronomy project, which has to process radio astronomy data from multiple telescopes, a task that requires a massive number of calculations. "The problem is that scientists go to the big computers to do their number crunching but then they use a different computer to look at their results," Haines says. "GPUs are letting us figure out not only how to process the data in real time, but how to let us look at the data while it is processing rather than at just the end."
Carnegie Mellon Computer Scientists Develop Method for Verifying Safety of Computer-Controlled Devices
Carnegie Mellon News (04/20/09) Spice, Byron
Carnegie Mellon University professors Edmund M. Clarke and Andre Platzer have developed a new method for identifying bugs in cyberphysical systems such as aircraft collision avoidance systems and high-speed train controls. The method has already discovered a bug in aircraft collision avoidance maneuvers that could have caused mid-air collisions, and has verified the soundness of the European Train Control System. "With systems becoming more and more complex, mere trial-and-error testing is unlikely to detect subtle problems in system design that can cause disastrous malfunctions," Clarke says. "Our method is the first that can prove these complex cyber-physical systems operate as intended, or else generate counterexamples of how they can fail using computer simulation." The method was used to analyze roundabout maneuvers in aircraft collisions, which are employed when two aircraft are on rapidly converging paths and involve both pilots turning right and then circling to the left until the two aircraft can safely turn right and continue on their original trajectories. The method found that when the aircraft approach at certain angles the roundabout maneuver actually creates a new collision course that the pilots may not be able to avoid. The method also could be used on other cyberphysical systems such as robotic surgery and nano-level manufacturing.
Researchers Use Brain Interface to Post to Twitter
University of Wisconsin-Madison (04/20/09) Meiller, Renee
University of Wisconsin-Madison biomedical engineering doctoral student Adam Wilson recently posted a status update on Twitter using only his thoughts. The message, "using EEG to send tweet," demonstrated a way for locked-in patients, those with no movement capabilities, to use brain-computer interfaces to communicate. Wilson is a part of a research effort to develop communications systems for people who cannot move but have normal brain function. "We started thinking that moving a cursor on a screen is a good scientific exercise," says UW-Madison professor Justin Williams, Wilson's advisor. "But when we talk to people who have locked-in syndrome or a spinal-cord injury, their No. 1 concern is communication." The new communication interface is based on brain activity related to changes in an object on a screen. The interface displays a keyboard on a computer screen, with each of the letters flashing individually. If a patient is focusing on a single letter and nothing is happening, there is no change in brain activity, but once that letter flashes, the brain recognizes that something is different and there is a momentary change in brain activity. Wilson says it is a slow process at first but that users improve as they use the interface, reaching up to eight characters per minute. The brain-based Twitter communication system represents one of the first uses of brain-computer interface techniques and Internet technologies, says Wadsworth Center research scientist Gerwin Schalk, who contributed to the project.
Lip-Reading Computers Can Detect Different Languages
University of East Anglia (04/09)
Computer scientists from the University of East Anglia (UEA) have developed a new lip-reading system that is capable of recognizing languages. "This is an exciting advance in automatic lip-reading technology and the first scientific confirmation of something we already intuitively suspected--that when people speak different languages, they use different mouth shapes in different sequences," says UEA professor Stephen Cox. "For example, we found frequent 'lip-rounding' among French speakers and more prominent tongue movements among Arabic speakers." UEA researchers taught the system to recognize English, French, German, Arabic, Mandarin, Cantonese, Italian, Polish, and Russian by modeling the lip motions made by a group of bilingual and trilingual speakers, and then having it identify which language was spoken by an individual speaker. The researchers now hope to gear the system to an individual's physiology and way of speaking. They say the system would be helpful for deaf people and law enforcement agencies.
Abstract News © Copyright 2009 INFORMATION, INC.
To submit feedback about ACM TechNews, contact: [email protected]
Change your Email Address for TechNews (log into myACM)