Welcome to the May 17, 2019 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."
|
|
|
UC Berkeley Grad Receives ACM Doctoral Dissertation Award
Association for Computing Machinery
Jim Ormond
May 15, 2019
ACM announced that University of California, Berkeley researcher Chelsea Finn will receive the 2018 ACM Doctoral Dissertation Award for her dissertation, "Learning to Learn with Gradients." In her thesis, Finn introduced algorithms for meta-learning that enable deep networks to solve new tasks from small datasets. Finn also demonstrated how the algorithms can be applied to fields such as computer vision, reinforcement learning, and robotics. Finn introduced a class of methods known as model-agnostic meta-learning (MAML) methods, which do not require computer scientists to manually design complex architectures. The new MAML methods have had a significant impact on the field and have been widely adopted in reinforcement learning computer vision, and other fields of machine learning. Finn will be recognized for her achievement at the annual ACM Awards Banquet on June 15 in San Francisco.
|
W&M Computing Mentorship for Girls Wins Global Service Award for Second Year in a Row
William & Mary
Adrienne Berard
May 15, 2019
The College of William & Mary's Society of Women in Computing (SWC) has been named recipient of the ACM Outstanding Community Service Award for the second consecutive year, for a mentorship designed for female middle-school students. Two years ago, SWC's board started organizing weekly lesson plans for a robotics mentorship with Williamsburg's Berkeley Middle School (BMS), including coding projects specifically geared for girls. Activities ranged from programming Harry Potter-style sorting hats to composing music and drawing shapes with microbots. Said William & Mary senior and SWC president Aparna Nagaraj, “We believe that it is our job to help improve our community by sparking and fostering an interest in computing, because we have the ability while many do not. Through that, we can reach a lot more girls and open doors for them.”
|
Driverless Electric Truck Starts Deliveries on Swedish Public Road
Reuters
Esha Vaish; Inze Filks; Anna Ringstrom
May 15, 2019
A driverless electric truck has started delivering freight daily along a public road in Sweden, an achievement described as a world first by vehicle developer Einride and logistics client DB Schenker. Einride CEO Robert Falck said, “This public road permit is a major milestone ... and it is a step to commercializing autonomous technology on roads.” The developer plans to have 200 vehicles dispatched by the end of next year. The T-Pod employs an Nvidia Drive platform for real-time visual data processing; a remote operator can coordinate and operate up to 10 T-Pods simultaneously. The vehicle can make short trips between a warehouse and terminal on a public road in an industrial area in Jonkoping, at a top speed of 5 kilometers (3.1 miles) an hour.
|
Microsoft Warns Wormable Windows Bug Could Lead to Another WannaCry
Ars Technica
Dan Goodin
May 14, 2019
Microsoft has warned a serious vulnerability in Windows Remote Desktop Services could be exploited to launch a WannaCry-level attack against Internet-connected computers. The pre-authentication vulnerability needs no human interaction to be exploited. Microsoft's Simon Pope called the bug "wormable," which means any future malware targeting the flaw could spread between infected computers. Microsoft has backported a just-released fix for the Windows 2003 and XP operating systems, which have lacked support for years. Windows 7, Windows Server 2008 R2, and Windows Server 2008 also are susceptible to the flaw.
|
Directors Using VR on Set to Find the Perfect Shot
New Scientist
Andrew Rosenblum
May 10, 2019
Technicolor has developed a virtual reality (VR) headset that allows film directors to instantly see what a scene will look like with computer-generated imagery added. Previously, directors had to send footage of human actors to render farms (groups of powerful computers) that would add graphical elements in a process that could take days to complete. The VR headset incorporates a stack of high-powered chips originally designed for gaming, which provide directors the opportunity to see immediately how a scene will look with the addition of computer-generated elements. The director can be immersed in the three-dimensional world of the scene and find the best camera angles. Independent filmmaker Kevin Margo said the technology can reduce the cost of visual effects up to 20% by trimming the expenses of lighting, and of the process to combine the elements into a single image.
|
Highest-Paying Jobs for the Class of 2019
Bloomberg
Shelly Hagan
May 15, 2019
Research by the Glassdoor review site listed data scientist as the highest-paid entry-level job of last year, with a median annual base salary of $95,000. Glassdoor analyst Amanda Stansell said, "There's such a high demand across industries for people with data science skills." Stansell predicted employers will continue to hire data scientists this year as they accrue data and mine it for actionable intelligence. Glassdoor's roster of the 25 highest-paying entry-level positions was dominated by technology, and software engineers were ranked second, with median annual base earnings of $90,000. The S&P 500 LinkUp Jobs Index estimated that the number of open information technology positions has risen more than 4% in the past year.
|
Brown Researcher Teaches Robots Handwriting, Drawing
Brown University (RI)
Kevin Stacey
May 15, 2019
Computer scientists at Brown University have created an algorithm that gives robots handwriting and drawing capability, which they called a step toward machines that can communicate and collaborate more seamlessly with humans. Brown's Atsunobu Kotani said, "Just by looking at a target image of a word or sketch, the robot can reproduce each stroke as one continuous action." The algorithm employs deep learning networks that analyze word or sketch images, inferring the likely sequence of pen strokes that generated them so it could replicate those images. When trained on a set of Japanese characters, not only could the algorithm reproduce those characters and their generative strokes with about 93% accuracy, but it could replicate dissimilar character types it had never before seen.
|
Image-Based Computer Model Reveals Finer Details of Tumor Blood Flow Behavior
Johns Hopkins Medicine Newsroom
May 6, 2019
Johns Hopkins Medicine researchers have developed a computer simulation of tumor behavior that accurately encapsulates finer blood-flow patterns, and how they engineer cancer growth. The model couples three-dimensional (3D) magnetic resonance microscopy and micro-computed tomographic imaging of tumor samples from mice with refined mathematical formulas to enable precise mapping of the blood vessel structure. The researchers developed a set of mathematical formulas based on data from previous studies, and their own, and converted the resulting data into an easy-to-visualize format. The result was a 3D rendering of the blood vessel network in each tumor, featuring a color-coded map of blood traffic through vessels that fed growth.
|
Using Smartphones to Sound Out Sign of Kids' Ear Infections
Associated Press
Lauran Neergaard
May 15, 2019
University of Washington scientists have developed a pediatric ear-infection diagnostic technique that uses a smartphone to detect fluid accumulation behind the eardrum. The technique involves cutting a piece of paper, folding it into a funnel configuration, and taping it around the smartphone's microphone and speakers. By focusing sound into the patient's ear canal with the funnel, the phone can emit birdlike chirps at a particular frequency, via an app. The microphone identifies sound waves bouncing off the eardrum, then the app analyzes the echo for signs of fluid buildup; the app sends the user a text indicating the likelihood middle-ear fluid is present. Testing the system on 98 ears in pre-surgical patients older than 18 months demonstrated the technique could detect fluid as well as or better than specialized acoustic testing technology.
|
Generating High-Quality Single Photons for Quantum Computing
MIT News
Rob Matheson
May 14, 2019
Researchers at the Massachusetts Institute of Technology (MIT) have created a method for generating, at room temperature, more single photons for carrying quantum information. A key challenge to developing a practical quantum computer is producing single photons with identical quantum properties, known as "indistinguishable" photons. Researchers previously found indistinguishability can be improved using emitters to funnel light through an optical cavity, which helps match their properties to the cavity. However, large cavities generate photons spontaneously, resulting in only a small fraction of photons staying in the cavity; smaller cavities extract higher percentages of photons, but they are of lower quality. The MIT team splits one cavity into two, each with a designated task: the small cavity handles the efficient extraction of photons, while the larger cavity stores them a bit longer to improve indistinguishability. The new coupled cavity generates photons with about 95% indistinguishability, compared to 80% for a single cavity, according to the researchers.
|
Federal Agencies Spend Millions to Hack Into Locked Phones
The Washington Post
Joseph Marks
May 13, 2019
U.S. federal law enforcement agencies are investing heavily in tools to bypass data encryption and access information on locked smartphones, without technology companies' participation. Immigration and Customs Enforcement (ICE) has contracted with Grayshift, vendor of a popular iPhone hacking kit. ICE has spent $1.2 million on iPhone hacking technology, and such efforts are being conducted amid rising anxiety about agencies' warrantless searches of phones and laptops at airports and other points of entry. Grayshift also has contracted with the Federal Bureau of Investigation and the Drug Enforcement Administration. Civil liberties advocates cited both ICE and Customs and Border Protection's arguments for asserting "near-unfettered authority to search and seize travelers' devices at the border … for purposes far afield from the enforcement of immigration and customs laws.”
*May Require Paid Registration
|
Researchers Train a Neural Network to Study Dark Matter
HPCwire
Kathy Kincade
May 14, 2019
Researchers at the National Energy Research Scientific Computing Center (NERSC), Berkeley Lab, Google Research, and the University of KwaZulu-Natal in South Africa have developed a new deep learning network, called CosmoGAN, which can create high-fidelity, weak gravitational lensing convergence maps through the use of generative adversarial networks (GANs). The team used the CosmoGAN generator network to produce convergence maps that are described by the same summary statistics as fully simulated maps. This very high level of agreement between convergence maps that are statistically indistinguishable from maps produced by generative models marks a step toward building emulators out of deep neural networks. Said NERSC researcher Mustafa Mustafa, "The idea of doing controllable GANs is essentially the Holy Grail of the whole problem that we are working on: to be able to truly emulate the physical simulators we need to build surrogate models based on controllable GANs."
|
|
