Association for Computing Machinery
Welcome to the September 7, 2011 edition of ACM TechNews, providing timely information for IT professionals three times a week.

Also, please download our new ACM TechNews iPhone App from the iTunes Store by clicking here and our new ACM TechNews iPad App by clicking here.


Computers Can See You--If You Have a Mug Shot
Wall Street Journal (09/03/11) Carl Bialik

Carnegie Mellon University (CMU) researchers recently presented data suggesting that facial recognition tools could identify individuals based on posed mug shots. The researchers demonstrated that, in principle, 33 percent of people photographed could be matched with a database of photos taken from Facebook. As part of the study, the researchers used images of 93 volunteers and compared them to Facebook photos of people on the CMU network. The results mean no one using facial-recognition software can claim "I can recognize any person in the U.S. at this very moment," says CMU's Ralph Gross. The problem is taking one image and comparing it to a wide set of images to find a single correct match. Comparing photos of just one person is easier and has achieved much more success. In a recent U.S. National Institute of Standards and Technology test, facial recognition software misidentified individuals in photographs just one percent of the time. Compared to Facebook images, closed-circuit (CC) TV images will probably be even more difficult to use with facial recognition systems, according to computer-vision experts. "Identifying faces in CCTV-quality images requires human experts," says University of Cambridge professor John Daugman.

Flash Memory That'll Keep on Shrinking
Technology Review (09/02/11) Katherine Bourzac

Researchers at the University of California, Los Angeles (UCLA) and Samsung have developed a type of flash memory that uses graphene and silicon to store information. The researchers say that the use of graphene could extend flash's viability far into the future and could lead to portable electronics that can store much more data than today's devices. "We're using graphene to help extend the capabilities of the conventional technology," says IBM Watson Research Center researcher Augustin Hong, who worked on the technology while at UCLA. The graphene flash memory prototypes use less power than conventional flash memory and they can store data more stably over time, even when miniaturized, according to the researchers. Graphene-based flash memory cells perform better due to the material's chemical structure and electrical properties, says UCLA professor Kang Wang, who led the research. Graphene also can hold much more charge than silicon without leaking, which is a common problem with conventional flash systems as the devices get smaller. Although the graphene-based devices have not been reduced to the size of some silicon-based devices, there are no known properties of graphene that would cause the performance to falter as the devices get smaller.

Online Activity Grows in a Similar Pattern to Those of Real-Life Networks
AlphaGalileo (09/01/11)

Accelerating growth patterns appear in the virtual world, according to Lingfei Wu, a researcher at the City University of Hong Kong. The phenomenon confirms that assigning user-chosen keywords to a piece of information to facilitate searches does not correlate in a linear way to the number of social media users using Internet tagging. Wu used the tagging behavior on the Flickr and Delicious social media sites to study the growing activity of online communities. Although the number of tags and the population fluctuates, communities have heterogeneity in individual tagging activity that remains constant over time, but differs across systems, according to Wu. The average individual activity will grow as the system expands and lead to the accelerating growth of overall activity. Such modeling of online activity growth could be used to predict the server capacity needs of social media sites on the basis of historical data. Wu plans to develop a unified model that explains the regularity governing the scaling up of both real-life systems and virtual communities.

Building Chips From Collapsing Nanopillars
MIT News (09/01/11) Larry Hardesty

A new technique demonstrated by researchers at the Massachusetts Institute of Technology's (MIT's) Research Laboratory of Electronics and the Engineering Agency for Science, Technology and Research (A*STAR) could serve as a fabrication method for nanoscale devices as they get smaller. The researchers say the technique could produce chip features only 10 nanometers across. The technique makes use of existing methods to deposit narrow pillars of plastic onto the surface of a chip. MIT professor Karl Berggren says the pillars do not have a rigid structure at that length scale and they just flop over. "It's more like trying to get a hair to stand up," he says. The researchers puzzled over the problem when it finally occurred to them that "if we can't end up beating it, maybe we can use it." Berggren and Lanzhou University student Huigao Duan have shown that when two pillars are very close to each other, they will collapse toward each other. In a forthcoming paper, Berggren, Duan, and Joel Yang show that by controlling the shape of isolated pillars, they can get them to collapse in whatever direction they choose.

To Clear Digital Waste in Computers, 'Think Green,' Johns Hopkins Researchers Say
Johns Hopkins University (09/01/11) Phil Sneiderman

Johns Hopkins University researchers have proposed using green solutions to solve digital waste data problems. The researchers, led by Ragib Hasan and Randal Burns, first determined what types of computer data qualify as waste. They note that many users may not even be aware that waste is piling up and impairing their computer's efficiency. "Like physical waste and trash, unwanted and unused data also pollutes the digital environment," the researchers say. "We propose using the lessons from real-life waste management in handling waste data." They developed a five-tier pyramid of options. At the top of the pyramid, the ideal option is to reduce the amount of waste data that flows into a computer by encouraging software developers to create programs that leave fewer unneeded files behind after a program is installed. Software makers also could reuse strings of code to have multiple functions, and some files could be recycled and used in other applications. Waste data that cannot be reused or recycled can be recovered to study patterns about historical trends. The researchers also suggest developing a digital landfill for data that cannot be reduced, reused, recycled or recovered.

Profiler at the Cellular Level
ETH Life (09/01/11) Peter Ruegg

Researchers at ETH Zurich and the Massachusetts Institute of Technology (MIT) have developed a diagnostic biological computer in human cells that can recognize certain cancer cells using logic combinations of five cancer-specific molecular factors. The researchers, led by ETH professor Yaakov Benenson and MIT professor Ron Weiss, developed a multi-gene synthetic circuit designed to distinguish between cancer and healthy cells, targeting the cancer cells for destruction. The circuit works by sampling and integrating five intracellular cancer-specific molecular factors and their concentration. When all five factors are present in a cell, the circuit makes a positive identification and starts the cell-destruction process. The researchers hope that their technique can be the basis for very specific anti-cancer treatments. They tested the gene network on cervical cancer cells and healthy cells, and found that the system only destroyed the cervical cancer cells. The five distinguishing factors are based on the cell's microRNA. The microRNA factors "are combined using logic operations such as AND and NOT, and the network only generates the required outcome, namely cell death, when the entire calculation with all the factors results in a logical TRUE value," Benenson says.

Robots Learn to Handle Objects, Understand New Places
Cornell Chronicle (09/01/11) Bill Steele

A team from Cornell University's Personal Robotics Laboratory is teaching a robot to find its way around new environments and manipulate objects, and machine learning is a key part of the project. "We just show the robot some examples and it learns to generalize the placing strategies and applies them to objects that were not seen before," says team leader and professor Ashutosh Saxena. "It learns about stability and other criteria for good placing for plates and cups, and when it sees a new object--a bowl--it applies them." The robot placed a plate, mug, martini glass, bowl, candy cane, disc, spoon, and tuning fork on a flat surface, on a hook, in a stemware holder, in a pen holder, and on several different dish racks. The robot surveyed its environment with a three-dimensional camera, randomly testing small volumes of space as suitable locations for placement. It placed objects correctly 98 percent of the time when it had seen the objects and environments previously, and 95 percent of the time when it did not. Saxena's team first developed a system to enable the robot to scan a room and identify its objects, training it on office and home scenes. The robot correctly identified objects about 83 percent of time in home scenes and 88 percent in offices.

Quantum Computer Chips Pass Key Milestones
New Scientist (09/01/11) Celeste Biever

University of California, Santa Barbara researchers have developed a quantum computer design based on superconducting electrical circuits that adds a more sustainable method to completing quantum calculations. Previously, quantum systems using photons or trapped ions as qubits have had the most success in solving advanced calculations, but the Santa Barbara researchers used tiny, superconducting wires to build the system. The researchers, led by Matteo Mariantoni, used a chip embedded with tiny loops of wire made from aluminum and rhenium. The researchers cooled the wires to a temperature very close to absolute zero, creating electrons coupled into "cooper pairs." The pairs were forced to resonate in unison, making them act like qubits. The researchers were able to move some quantum information from one qubit wire to another, thus entangling the qubits. The researchers also were able to use the system to develop a Toffoil OR phase gate, which is an important step towards building codes that perform quantum error correction. "The beautiful thing about a solid circuit is that it's something you can write using lithographic technology," says University of Queensland researcher Andrew White.
View Full Article - May Require Free Registration | Return to Headlines

Mobile Phone Data in Haiti Improves Emergency Aid
Karolinska Insitutet (08/31/11) Katarina Sternudd

Researchers at the Karolinska Institutet and Columbia University developed a program to monitor the daily movements of 2 million mobile phone users in Haiti to track cholera outbreaks. "We rapidly received mobile phone data and within 12 hours we were able to send out analyses describing which areas had received people from the cholera outbreak zone in order to provide information on areas at potentially increased risk of new outbreaks," says Karolinska's Linus Bengtsson. The researchers launched the project after the devastating earthquake in January 2010, and estimated that more than 600,000 people left Port-au-Prince within 19 days of the disaster. Although the researchers have shown that the movements of the mobile phones match those of a United Nations-led study conducted during a stable phase six months after the earthquake, their analysis differed from estimated migration patterns that were used during the initial phase of relief response. "We believe that the method can bring about important improvements in humanitarian relief and development cooperation," Bengtsson says.

UC Davis Researchers Develop Computer Model for Testing Heart-Disease Drugs
UC Davis News & Information (08/31/11) Karen Finney

University of California, Davis researchers have developed a computer model to test the effects of medication for arrhythmia before they are used on patients. The tool will help researchers study anti-arrhythmia medications early in the drug-development process. The researchers, led by Davis' Colleen Clancy, started with existing models that simulate the behavior of heart cells during normal heart rhythms and arrhythmia. Clancy then created mathematical formulas to describe the interactions of drugs, such as flecainide and lidocaine, with their specific cell targets. The researchers used the models to predict the effects of the drugs on cardiac rhythms in a three-dimensional virtual heart. "An additional benefit of this model is that we could see exactly how flecainide causes arrhythmia," Clancy says. A major benefit of the computer-based drug-testing approach is that it enables researchers to simultaneously follow many processes, according to Davis researcher Jonathan D. Moreno. To confirm the results, the researchers treated rabbit hearts with the two drugs and found that the simulation correctly predicted the heart rates and concentrations of the adverse effects.

Citizen Scientists Use Games to Design Protein and RNA
SmartPlanet (08/30/11) Christina Hernandez

Untrained online gamers are more talented than the best computer programs when it comes to protein-folding problems, says Carnegie Mellon University professor Adrien Treuille in an interview. Thousands of citizen scientists are now using his games Foldit and EteRNA to design protein and RNA strands. Treuille says he turned to computer games to help people understand movement, shapes, and structures because he wanted to take graphics and simulation to the next level. Treuille says the games enable people to experience protein or an RNA, play with it like a toy, and possibly devise answers for scientific problems. "We're allowing people to do what humans do best, which is to find the patterns," Treuille says. He notes that the difference between humans and computer programs has been even more dramatic in predicting how RNAs are going to fold. "It's fair to say that some of the best empirical RNA designers in the world are our players," Treuille points out. He considers the elite players to be sophisticated scientists, and believes real medical advances could be based on what these players are doing.

Birthing Games: Assistive Technology Lab Delivers Serious Perinatal Games
CITRIS Newsletter (08/29/11) Gordy Slack

University of California, Santa Cruz graduate student Alexandra Holloway and her colleagues have created several games that help expectant mothers and their partners prepare for childbirth and infant care. In one game, users assist a pregnant virtual woman, Amy, who has just begun her first stage of labor. The player must help Amy maintain her energy and remain cognitively and physically strong. The partner must monitor the mother's needs and respond appropriately or the doctor will be called to perform a Caesarean section. In testing, the game was played by 50 participants who took a test about child birth before and after playing it. The study found that the participants had a significant test score improvement after playing the game. Another game, which is played over a period of days or weeks, involves a simulated pregnancy that has just entered its third trimester. The player must respond to the mother's water breaking, her needs, and the suggestions given by the nurse, midwife, or doctor. A third game is designed to help expecting parents learn the difficulties some newborns can have latching on for breastfeeding.

Accelerating Discovery by Outsourcing the Mundane: An Interview With Ian Foster
HPC in the Cloud (08/29/11) Wolfgang Gentzsch

University of Chicago professor and Computation Institute director Ian Foster says in an interview that software as a service (SaaS) can play a crucial role in expediting discovery. "If we focus on hardware alone, we miss the real benefits of the large-scale outsourcing and consequent economies of scale that cloud [computing] is about," he argues. Foster contends that creativity is being stifled by the need to establish and run big data collection, management, analysis, sharing, and archiving processes, and he sees no reason why researchers cannot outsource such mundane functions to a SaaS provider. Supplying research management processes via SaaS is the goal of the Globus Online project, which aims to "make the discovery potential of massive data, exponentially faster computers, and deep interdisciplinary collaboration accessible to every one of the million or more professional researchers worldwide--and to the billions of potential citizen scientists," according to Foster. He says the appeal of Globus Online lies in its design to address the needs of small science projects, and notes that in its first eight months of operation the project has accumulated 1,600 registered users, nearly 1 petabyte of data, and more than 30 million files.

Abstract News © Copyright 2011 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