Association for Computing Machinery
Welcome to the September 10, 2012 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).

HEADLINES AT A GLANCE


Robots at War: Scholars Debate the Ethical Issues
Chronicle of Higher Education (09/10/12) Don Troop

Lethal autonomous systems are gradually penetrating battlefield operations, and Georgia Tech professor Ronald C. Arkin predicts the advent of robots that are ethically superior to human soldiers. "I'm taking about ... very specific machines for certain tasks that will work alongside human war fighters to carry out particular types of operations that humans don't do particularly well at, such as building-clearing operations," he says. Arkin has created algorithms for an ethical governor to direct such machines to either shoot or hold their fire in accordance with international rules of law. The algorithms came out of a project funded by a three-year grant from the U.S. Army Research Office to produce an artificial conscience to guide robots in the battlefield independent of human control. Arkin says the decision-making architecture could potentially support the creation of ethically superior robotic fighters in as little as one to two decades. The independence of autonomous robots also could be an asset in situations where the enemy severs the communications link between controller and drone. University of Sheffield professor Noel Sharkey counters that a robot-sensing system cannot attain a level of sophistication to discriminate between combatants and civilians within Arkin's proposed timeline.


Stanford Announces 16 Online Courses for Fall Quarter
Stanford Report (CA) (09/07/12) Lisa Lapin

Stanford University launched 16 new online courses and two new home-grown platforms for interactive learning this fall. Stanford's new online courses cover topics in computer science, mathematics, linguistics, science writing, sociology, and education. The two platforms each have distinct features and capabilities, such as video lectures, discussion forums, peer assessment, problem sets, quizzes, and team projects. Class2Go is a new open source platform developed by Stanford engineers that will host "An Introduction to Computer Networks" and a course on solar cells. The other new platform, called Venture Lab, will host "Technology Entrepreneurship," which attracted 37,000 students when it was first offered last spring, as well as other courses. Venture Lab, developed by Stanford's Amin Saberi, was designed for use by students working in teams. Meanwhile, the Coursera online learning platform will host nine Stanford courses this fall, including "Writing in the Sciences" and "Human-Computer Interaction." Other online courses include those on machine learning, cryptography, probabilistic graphical models, organizational analysis, and finance.


Artificial Intelligence, Powered by Many Humans
Technology Review (09/10/12) Tom Simonite

University of Rochester researchers have developed Chorus, an approach to virtual personal assistants that creates a smart artificial chat partner from small contributions from many crowdsourced workers. During testing, Chorus was asked for travel advice and the system showed that it could be smarter than any one individual in the crowd, because multiple people were contributing to its responses. "It shows how a crowd-powered system that is relatively simple can do something that [artificial intelligence] has struggled to do for decades," says Rochester professor Jeffrey Bigham. The researchers aimed to find a new way to increase the power of crowdsourcing, which is traditionally limited to simple, isolated tasks. "What we're really interested in is when a crowd as a collective can do better than even a high-quality individual," Bigham says. As part of the Chorus system, any new chat updates from the user are passed along to many crowd workers, who are asked to suggest a reply. The suggestions are then voted on by crowd workers to determine which one will be sent back to the user. The final step creates a working memory that ensures that Chorus' replies reflect the history of the conversation.


Who’s the Most Influential in a Social Graph?
Georgia Tech News (09/07/12) Jason Maderer

Georgia Tech researchers say they have developed an algorithm that quickly determines betweenness centrality for streaming graphs. They say the algorithm also can identify influencers as information changes within a network. "Our algorithm stores the graph’s prior centrality data and only does the bare minimal computations affected by the inserted edges," says Georgia Tech professor David Bader. In some situations, Bader says the software can compute betweenness centrality more than 100 times faster than conventional methods. He notes advertisers could use the software to identify which celebrities are most influential on social media during product launches. "Despite a fragmented social media landscape, data analysts would be able to use the algorithm to look at each social media network and mark inferences about a single influencer across these different platforms," Bader says.


Mind Control Will Shape Future of Gaming and Cell Phones
University of Alabama Huntsville (09/06/12) Joyce Anderson Maples

In an interview, University of Alabama at Huntsville professor Jason T. Cassibry shares his thoughts about the new technological advances on the horizon. He says that in the next few years gamers will use their minds to control and communicate with gaming stations. "Perhaps transmitters will communicate back and generate gaming experiences in the player's mind without requiring a screen," Cassibry says. Cell phones will have transparent, lightweight, foldable screens, and eventually users will be able to control the devices with their minds. The military and law enforcement are already using drones, but advancements in robotics engineering technology will pave the way for personal robots. "Gradually, more and more people's lives will be enhanced by robots that will take care of simple and, eventually, more sophisticated tasks," Cassibry says. He also believes exascale computing will allow for more precise predictions of weather forecasts. "The interactions among the sun, oceans, and variations in terrain, cosmic rays, and precipitation will be better understood and will be able to leverage the enormous computing power to enable precise predictions of weather," Cassibry says.


Cheetah Robot 'Runs Faster Than Usain Bolt'
BBC News (09/06/12)

Boston Dynamics, with support from the U.S. Defense Advanced Research Projects Agency (DARPA), has developed Cheetah, a robot that recently set a new world record for legged robots, reaching 28.3 miles per hour. In comparison, current 100-meter dash world record holder Usain Bolt's top speed is 27.78 miles per hour. DARPA says the goal of the Cheetah project is to "more efficiently assist war fighters across a greater range of missions." The robot is powered by a hydraulic pump, and it passed an earlier record of 18 miles per hour recorded in February. "The Cheetah had a slight advantage over Bolt as it ran on a treadmill, but most of the power Cheetah used was to swing and lift its legs fast enough, not to propel itself forward," DARPA says. DARPA program manager Gill Pratt notes "our Cheetah bot borrows ideas from nature's design to inform stride patterns, flexing and unflexing of parts like the back, placement of limbs, and stability. What we gain through Cheetah and related research efforts are technological building blocks that create possibilities for a whole range of robots suited to future Department of Defense missions."


Math Tree May Help Root Out Fraudsters
University of Alberta (09/06/12) Jamie Hanlon

Researchers at the University of Alberta, University of Connecticut, and University of California-Merced recently outlined the connection linking fraud cases and the algorithm designed by Swiss mathematician Jakob Steiner, known as Steiner's Tree. Alberta researcher Ray Patterson says that for crimes such as fraud, the fewer players in the scheme, the more likely it will be realized. Maintaining a small group of players also is what connects it to the Steiner tree, and Patterson says that by analyzing various connecting social networks--email, Facebook, etc.--learning the who, what, and how of the crime can be distilled into numbers. The researchers explored how networks such as phone calls, business partnerships, and family relationships are used to form essential relationships in a fraud investigation. Once unnecessary links are removed and false leads are extracted, the remaining connections are most likely the best suspects. Finding the shortest connection between the criminals and the crime is the key to the Steiner tree, according to Patterson. "If you can reduce your legwork by even 20 percent, that has massive manpower implications," he notes.


U.S. States Intention to Retain Control of the Internet
Techworld (09/05/12) Sophie Curtis

Although the Internet Corporation for Assigned Names and Numbers (ICANN) is currently responsible for the coordination of Internet Protocol (IP) addresses, the assignment of address blocks, the maintenance of registries, and the management of the top-level domain name space, it has been suggested that the United Nation's International Telecommunications Union (ITU) could perform at least some of these duties. In December, the ITU will host a conference to which representatives from 178 countries have been invited to review the International Telecommunications Regulations, which describe how traffic should flow between telecommunications networks in different countries. Russia is planning to submit a plan that calls for the ITU to be responsible for the allocation of IP addresses and the determination of necessary requirements. However, the United States is worried that "proposals by some other governments could lead to greater regulatory burdens being placed on the international telecom sector," according to its own proposal. "The United States also believes that the existing multi-stakeholder institutions, incorporating industry and civil society, have functioned effectively and will continue to ensure the health and growth of the Internet and all of its benefits," says U.S. ambassador to the conference Terry Kramer.


A Step Forward for Ultrafast Spintronics
Uppsala University (09/05/12) David Naylor

Uppsala University researchers have demonstrated how spin information can be transmitted using spin currents at terahertz speeds and how spin currents can be generated and transferred from one nanoscale metal layer to another in less than a picosecond. The method involves using ultra-short light pulses to create spin currents that move super-fast through the nanostructured layers and transfer the spin. "Since the generated spin currents transport the spin magnetic moment in just a few femtoseconds, new technology is needed to detect them," says Uppsala researcher Peter Oppeneer. "We succeeded to do so by using ultra-short x-ray flashes with a pulse length of just a few femtoseconds." Spin magnetic moments were quantified using magnetic x-ray spectroscopy, which makes it possible to perceive the spin dynamic processes that up to now have been invisible. The Uppsala researchers have devised a basic theory for these superdiffusive spin currents, which have been found to be the cause of ultra-fast spin dynamics. "A distinguishing feature of super-fast spin currents is that they are caused by non-equilibrium processes, which is also the reason for their higher speed," Oppeneer says.


New Program Joins Computer Science and Design Experts at UW, Tsinghua University
UW News (WA) (08/30/12) Hannah Hickey

The University of Washington (UW) recently hosted the first World Lab Summer Institute, which brought together UW and Tsinghua University students with the goal of developing ways that technology could be used to address global issues relating to health, the environment, and education. The students developed prototype applications, including a social networking tool to donate materials and view recycled artworks, an armband that prompts people to incorporate micro-exercise in their day, and a tool to help parents be more closely involved in many aspects of their child's education. The projects were developed by teams of four students, mainly divided between Chinese and Americans, and with an equal blend of design and technical knowledge. Cultural differences influenced how the students developed their ideas and how they chose to present the apps. For example, the UW students saw a dandelion as a weed, while the Chinese students saw it as a sign of hope. The program aimed to combine these two differing world views during the development of the apps. "I hope that this becomes a permanent research institute that works jointly on these types of grand-challenge problems," says UW professor and World Lab founder James Landay.


Unsure Robots Make Better Teachers Than Know-Alls
New Scientist (08/31/12) Douglas Heaven

The results of an experiment by researchers at the University of Tsukuba show that young students performed best when a robot appeared to learn from them. Shizuko Matsuzoe and Fumihide Tanaka also found that this made the children more likely to want to continue learning with the humanoid Nao robot. For the experiment, Matsuzoe and Tanaka had 19 children aged four through eight interact with the robot in a learning game of drawing the shape that corresponded to an English word such as circle, square, or heart. The robot could either draw the correct shape or make mistakes and act as if it did not know the answer, and the children could teach the robot how to draw it correctly by guiding its hand. The robot either learned the English word or continued to make mistakes. "Anything that gets a person more actively engaged and motivated is going to be beneficial to the learning process," says Andrea Thomaz, director of the Georgia Institute of Technology's Socially Intelligent Machines lab. "So needing to teach the robot is a great way of doing that."


Eye Movements Could Help in Diagnosis of Neurological Disorders
USC News (08/30/12) Katie Dunham

University of Southern California (USC) researchers have developed a method for detecting several neurological disorders by studying a person's eye movements. Attention Deficit Hyperactivity Disorder (ADHD), Fetal Alcohol Spectrum Disorder (FASD), and Parkinson's disease each involve ocular control and attention dysfunctions, which can be identified through an evaluation of how people move their eyes while they watch TV. As part of the study, participants were asked to watch TV clips for 20 minutes while their eye movements were recorded. Their eye-tracking data then was combined with eye-tracking data from healthy volunteers, and a computational model of visual attention extracted 224 quantitative features. The researchers then used new machine-learning techniques to identify critical features that stood out in ADHD, FASD, and Parkinson's disease patients. The technique was able to identify older adults with Parkinson's disease with 89.6 percent accuracy, and children with either ADHD or FASD with 77.3 percent accuracy. The researches say their method provides an easily deployed, low-cost, high-throughput screening tool. "For the first time, we can actually decode a person’s neurological state from their everyday behavior, without having to subject them to difficult or time-consuming tests," says USC professor Laurent Itti.


Abstract News © Copyright 2012 INFORMATION, INC.
Powered by Information, Inc.


To submit feedback about ACM TechNews, contact: technews@hq.acm.org
Current ACM Members: Unsubscribe/Change your email subscription by logging in at myACM.
Non-Members: Unsubscribe


About ACM | Contact us | Boards & Committees | Press Room | Membership | Privacy Policy | Code of Ethics | System Availability | Copyright © 2014, ACM, Inc.