Welcome to the March 23, 2018 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).
To view "Headlines At A Glance," hit the link labeled "Click here to view this online" found at the top of the page in the html version.
The online version now has a button at the top labeled "Show Headlines."
|
|
|
Golden Touch: Next-Gen Optical Disk to Solve Data Storage Challenge
RMIT News
March 22, 2016
Researchers from RMIT University in Australia and Wuhan Institute of Technology in China have demonstrated a high-capacity optical disk that can store data securely for more than six centuries, using gold nanomaterials. The disk possesses as much as 10 TB in memory capacity, and requires far less cooling than current storage media. The team developed a nanoplasmonic hybrid glass matrix, incorporating gold nanorods into a hybrid glass composite known as organic modified ceramic. The researchers say they opted for gold because it is robust and highly durable. Gold nanoparticles enable information to be recorded in five dimensions--the three dimensions in space, in addition to color and polarization. The chip is fabricated via a sol-gel process that employs chemical precursors to produce ceramics and glasses with better purity and homogeneity than conventional processes. "Our technique can create an optical disk with the largest capacity of any optical technology developed to date," says RMIT professor Min Gu.
|
Powerful New Algorithm Is a Big Step Towards Whole-Brain Simulation
Singularity Hub
Shelly Fan
March 21, 2018
An international team has overhauled the structure of a popular simulation algorithm, cutting computing time and memory consumption. "With the new technology we can exploit the increased parallelism of modern microprocessors a lot better than previously, which will become even more important in exascale computers," says Jakob Jordan of the Julich Research Center in Germany. On Germany's JUQUEEN supercomputer, the algorithm ran 55 percent faster than previous models on a random neural network, mainly due to a streamlined data transfer scheme. For example, at a network size of 500 million neurons, simulating one second of biological events took about five minutes of JUQUEEN runtime using the new algorithm, while its predecessor took six times as long. "The novel technology profits from sending only the relevant spikes to each process," the researchers note. They also say since computer memory is now uncoupled from the size of the network, the algorithm is poised to tackle brain-wide models.
|
Illusory Motion Reproduced by Deep Neural Networks Trained for Prediction
National Institute for Basic Biology
March 20, 2018
Researchers at the National Institute for Basic Biology (NIBB) in Japan have successfully replicated illusory motion by predictive deep neural networks (DNNs). The networks are founded on predictive coding theory, which assumes that the internal models of the brain anticipate the visual world at all times and that errors between the prediction and the actual sensory input further refine the models. The DNNs were trained with natural scene videos of motion from the viewer's perspective, and the motion prediction ability of the computer model was confirmed using a rotating propeller in unlearned videos and the "Rotating Snake Illusion." The model accurately predicted the magnitude and direction of motion of the rotating propeller in the unlearned videos, and represented the rotational motion for illusion images that were not in motion. "This research supports the exciting idea that the mechanism assumed by the predictive coding theory is a basis of motion illusion generation," says NIBB professor Eiji Watanabe.
|
Bipedal Robot Uses Jet-Powered Feet to Step Over Large Gaps
IEEE Spectrum
Evan Ackerman
March 19, 2018
Researchers at Guangdong University of Technology's School of Automation in China are experimenting with the use of small ducted fans embedded in the feet of a bipedal robot that can change the machine's center of gravity and help it balance as it takes giant steps over a wide gap it would normally be unable to traverse. At the end of each foot is a lightweight ducted fan jet engine that can release up to 2 kilograms of thrust, nearly 33 percent of the weight of the entire robot. "Our idea was utilizing the external force of the jets to maintain the balance of the robot," says Guangdong University of Technology professor Zhifeng Huang. The team suggests this method could be adapted to other scenarios in which robots need help controlling their balance when confronted with large obstacles that they must step over, or preventing them from toppling when traversing uneven terrain.
|
Physicists Reveal Material for High-Speed Quantum Internet
MIPT News (Russia)
March 21, 2018
Researchers from the Moscow Institute of Physics and Technology in Russia say they have "rediscovered" a material upon which a practical ultrahigh-speed quantum Internet can be based. They studied the physics of electroluminescence of color centers in silicon carbide, formulating a theory of single-photon emission upon electrical injection. A color center is a point defect in the lattice structure of silicon carbide that can discharge or absorb a photon at a wavelength to which the material is transparent in the absence of defects. The team demonstrated how a single-photon-emitting diode based on silicon carbide can be enhanced to emit up to several billion photons every second, which is required to deploy quantum cryptography protocols at data transfer rates on the order of 1 Gbps. Silicon carbide-based single-photon sources also interoperate with complementary metal-oxide semiconductor (CMOS) technology. This makes silicon carbide by far the most promising material for constructing practical ultrawide-bandwidth data communication lines that are unconditionally secure.
|
Google Is Working on Its Own Blockchain-Related Technology
Bloomberg
Olga Kharif; Mark Bergen; Joe Mayes
March 21, 2018
Google researchers are developing blockchain-related technology to support its cloud business and ward off competition from emerging startups. Google insiders recently noted the cloud business is a natural place for blockchain-related services. To construct its ledger, Google has explored technology from the Hyperledger consortium, but it could choose another type that may be less difficult to scale to run millions of transactions. Google's Sridhar Ramaswamy says his division has a "small team" investigating blockchain, but the current core technology cannot accommodate many transactions quickly. One possible application of blockchain technology by Google could entail reassuring its customers that their information is shielded when stored on the giant network of computer servers that power Google's cloud services. "Like many new technologies, we have individuals in various teams exploring potential uses of blockchain but it's way too early for us to speculate about any possible uses or plans," says a company spokesperson.
|
Programming Languages Can Be Hard to Grasp for Non-English Speakers. Step Forward, Bato: A Ruby Port for Filipinos
The Register (UK)
Shaun Nichols
March 21, 2018
Filipino developer Joel Bryan Juliano says he built Bato, a handmade Ruby port, so Tagalog speakers can learn programming fundamentals without having to be English-proficient. Structure was a major obstacle for Juliano when choosing a language to port, since so many programming environments are based on English. Ruby's structure mirrors that of Tagalog, Juliano says, because "focusing on Ruby with Tagalog syntax feels more natural to Filipinos." Juliano published Bato on GitHub with an interactive console so Internet users can try out code directly from their browsers. Juliano notes he ultimately wants to expand Bato from a GitHub project to a fully-fledged educational platform that will include textbooks and low-cost notebooks. "I'd like to provide a way for Filipinos to learn to code, and work in the [information technology] industry, and save them from the abusive call center companies," he says.
|
Second Annual Women in Data Science Conference Showcases Research, Explores Challenges
MIT News
Scott Murray
March 19, 2018
The Massachusetts Institute of Technology (MIT) Institute for Data, Systems, and Society (IDSS) recently convened 200 students, industry professionals, and academic leaders at the second annual Women in Data Science conference in Cambridge, MA. The Institute co-hosted the event with Harvard University's Institute for Applied Computational Science and Microsoft Research New England, in partnership with Stanford University. MIT's Regina Barzilay discussed ways in which data science and machine learning are improving cancer research; her research uses natural language processing and deep neural networks to organize and analyze previously unused patient data to shed light on treatment outcomes. MIT's Tamara Broderick spoke about developing a modern toolbox for data analysis, with tools that run on streaming data and leverage modern, distributed computing. Participants addressed challenges such as bias, transparency, security, and privacy, but concurred that data science can significantly impact social challenges.
|
A New Way to Gauge the Growth of Nanowires
Argonne National Laboratory
Jared Sagoff
March 19, 2018
Researchers from the U.S. Department of Energy's Argonne and Brookhaven National Laboratories have discovered the formation of two types of defects in individual nanowires made of indium gallium arsenide. The nanowires could be useful for optoelectronic applications such as fiber optic relays, but their performance is altered by defects that can change optical and electronic properties. Scientists hope to control these defects to improve the functionality of future optoelectronics, including materials that could manipulate quantum information. The team, which included researchers from Northwestern University and two European universities, observed one type of defect caused by strain, which affects the entire nanowire and prevents it from growing straight. The second type of defect is a stacking fault, which occurs close to the atomic level as individual planes of atoms are laid down to lengthen the nanowire. Using Bragg ptychography to observe the defects, the researchers devised a method to see the nanowire within its operating environment.
|
Scientists Use Big Data to Awaken Stem Cells and Heal Brain Tissue
Interesting Engineering
Sibel Nicholson
March 15, 2018
Researchers at the University of Toronto in Canada reported in a study that they have combined single-cell genomic technology with machine learning to yield knowledge on brain stem cells that could eventually help the brain heal itself. "This research tells us more about how adult neural stem cells are formed in the brain, which is vital for developing future medical treatments for brain and nervous system injuries," says University of Toronto professor Gary Bader. Single-cell genomics enabled the scientists to identify when cortical stem cells would get their lifetime identity, and they believe this information can deliver insights on how to invigorate stem cells to support tissue repair. The team wants to use stem cells in the body to heal tissue from inside, and with single-cell genomics they gain understanding of which genes are activated in many individual tissue cells. By using this technology, they were able to pick out extremely rare stem cells.
|
New Computing Device Would Let Microprocessors Go 'All Out'
University of Rochester NewsCenter
Bob Marcotte
March 15, 2018
A researcher at the University of Rochester has proposed a wholly new electrically reconfigurable logic gate to overcome the problems of heat transfer innate in traditional computer architecture. Mohammad Kazemi says the gate is designed to combine processing and memory functions in individual magnetic nanodevices, instead of partitioning the functions between different pieces of hardware. The device also taps the spin of electrons, as well as their electrical charge, for processing, and employs bounded switching to reconfigure the gate for different logic operations. Kazemi also notes the gate integrates these functions in a physical "footprint" that is 10 times smaller than other "state-of-the-art" gates. "This has significant potential for enhancing the performance of microprocessors by orders of magnitude," he says. Kazemi believes the device will better fulfill the requirements of emerging data processing and learning applications because it can deliver faster and more energy-efficient computational capabilities.
|
|
