Welcome to the September 16, 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."
|
|
Noise-Canceling Breakthrough to Accelerate Quantum Computer Tech
Silicon Republic Colm Gorey September 16, 2019
Computer scientists at Dartmouth College and the Massachusetts Institute of Technology have found a way to reduce unwanted noise around quantum computers, to make their computations more precise. The researchers developed and performed a laboratory test that detected and characterized a class of complex "non-Gaussian" noise processes, which are hard to predict and which a quantum computer can confuse with the quantum signal it is supposed to process. Dartmouth's Lorenza Viola said the experiment was "the first time that a detailed, frequency-resolved characterization of non-Gaussian noise has been able to be done in a lab with qubits.” Viola added, “This result significantly expands the toolbox that we have available for doing accurate noise characterization and therefore crafting better and more stable qubits in quantum computers."
|
Hackers Exploiting Platform-Agnostic Flaw to Track Mobile Phone Locations
Ars Technica Dan Goodin September 12, 2019
Hackers are taking advantage of a mobile phone vulnerability to secretly track users' whereabouts, AdaptiveMobile Security researchers disclosed. The Simjacker hack targets an interface designed for exclusive use by cellular carriers to communicate directly with SIM cards in subscribers' phones and deliver specialized services. Simjacker transmits commands that track location and acquire the phones' International Mobile Equipment Identity code; attackers also might instruct phones to make calls, send texts, or execute other commands. Security firm Trail of Bits CEO Dan Guido said, "This attack is platform-agnostic, affects nearly every phone, and there is little anyone except your cell carrier can do about it."
|
A Smart Artificial Hand for Amputees Merges User, Robotic Control
Ecole Polytechnique Fédérale de Lausanne Hillary Santuary September 11, 2019
Researchers at Ecole Polytechnique Fédérale de Lausanne (EPFL) in Switzerland are developing new approaches for improved control of robotic hands for amputees that combine individual finger control and automation for improved grasping and manipulation. The team successfully tested the proof-of-concept device on three amputees and seven healthy subjects. The technology combines neuroengineering and a robotic system that allows the prosthetic hand to take hold of objects and maintain contact with them for robust grasping. The system learns how to decode user intention and translates it into finger movement of the prosthetic hand. Said EPFL researcher Silvestro Micera, "Our shared approach to control robotic hands could be used in several neuroprosthetic applications such as bionic hand prostheses and brain-to-machine interfaces, increasing the clinical impact and usability of these devices."
|
Israeli Researchers Develop AI-Based Cancer Treatment
The Jerusalem Post Leon Sverdlov September 11, 2019
Technion-Israel Institute of Technology researchers have developed an artificial intelligence (AI)-based technique for mapping receptors on cancer cells. The technique derives molecular data from images of breast cancer biopsies that received hematoxylin and eosin staining, extracting biological properties to help pathologists match patients with a therapy that will block receptors on the cell membrane and hinder tumor growth. Technion's Gil Shamai and Ron Kimmel said the deep learning technique successfully identifies cancer's unique morphological signature within tissue. Shamai said to feed the AI system the necessary cancer information, "We have written software code to scan network sources and automatically download thousands of biopsy samples and the relevant medical information approved for research."
|
GymCam Tracks Exercises That Wearable Monitors Cannot
Carnegie Mellon University Human-Computer Interaction Institute September 12, 2019
Researchers at Carnegie Mellon University's Human-Computer Interaction Institute (HCII) found that a vision-based stationary camera offers a better way to monitor gym exercises than wearable sensors. The GymCam system can identify types of exercise and reliably tally repetitions. GymCam only requires motion data, so the camera feed can be simplified to pixel-by-pixel changes, removing identifiable faces. HCII's Rushil Khurana said GymCam also overcomes single-camera systems' inability to view a person's entire body—all it needs to detect exercise is to observe any body part moving repetitively. HCII's Mayank Goel added that the GymCam algorithm can function on a smartphone, allowing users to record and track at-home workouts.
|
Tracking Coordinated Disinformation Campaigns Online Made Easier with New BotSlayer Tool
Indiana University Kevin Fryling September 12, 2019
Indiana University researchers have developed a software program that scans social media in real time to detect evidence of automated Twitter accounts pushing messages in a coordinated manner. BotSlayer gives groups and individuals of any political affiliation the power to detect coordinated disinformation campaigns in real time. BotSlayer was created in part based on feedback from political and news organizations asking for similar tools to be made faster, more powerful, and more user-friendly. The system relies on an "anomaly detection algorithm" to report sudden surges of trending activity likely driven by bots. In addition to detecting trends, BotSlayer can also instantly generate a "network map" that shows how a particular topic is spreading over time. A bot score is assigned to each user in the network, providing a way to see the most influential accounts in the conversation, whether they are real or fake.
|
Popular Mobile Games Can Detect Cognitive Decline
University of Kent Michelle Ulyatt September 12, 2019
Researchers at the University of Kent in the U.K. have found that popular mobile phone games could be used as a tool to help doctors identify early signs of cognitive decline, including the onset of serious conditions like dementia. The researchers studied the link between patterns of tap, swipe, and rotational gestures during mobile game play, and the users' cognitive performance; the results showed the speed, length, and intensity of these motions correlates with brain function, offering clues about the individuals' overall brain health. Said Kent researcher Jim Ang, "We're now working to design an algorithm which can carry out automatic monitoring of individuals' cognitive performance while playing these games."
|
Smile-to-Pay: Chinese Shoppers Turn to Facial Payment Technology
The Guardian September 4, 2019
China is embracing facial payment technology, which allows consumers to purchase goods simply by posing in front of point-of-sale (POS) machines equipped with cameras after linking an image of their face to a digital payment system or bank account. Similar facial recognition software is already widely used by the government to monitor citizens, although authorities have been criticized for using it to crack down on dissent. Despite concerns over data security and privacy, many consumers seem unfazed by facial payment technology. Alipay—the financial arm of Alibaba—already has facial payment devices in place in 100 Chinese cities.
|
Intel Server-Grade CPUs Impacted by New NetCAT Attack
ZDNet Catalin Cimpanu September 10, 2019
Researchers at Vrije University in the Netherlands described a new exploit targeting Intel central processing units (CPU) that support Data-Direct I/O Technology (Intel DDIO) and Remote Direct Memory Access (RDMA) features. The NetCAT side-channel attack against remote, networked systems is facilitated when both features are enabled, allowing hackers to deduce certain kinds of data being processed in the CPU's cache. The technique looks for slight variations in how long the CPU needs to process data, then guesses what data may have been processed. The Intel DDIO and RDMA features affect the NetCAT hack through network packets sent to a computer's network card; Intel DDIO permits peripherals to directly access write data inside the cache, rather than random-access memory, while RDMA can accelerate networking speed and throughput to make the attack more efficient. A patch for this exploit is not yet available.
|
New AR Head-Mounted Display Offers Unrivaled Viewing Experience
University of Cambridge (UK) September 13, 2019
An augmented reality head-mounted display (HMD) developed by researchers at the University of Cambridge in the U.K. and the Huawei European Research Center in Germany provides a realistic three-dimensional (3D) viewing experience free of nausea or eyestrain. The HMD features a scalable eye-box and a field of view extended by 36 degrees, displaying images on the user's retina via pixel-beam scanning to ensure the image remains in focus irrespective of an object’s distance from the wearer. The device employs partially reflective beam splitters to generate an additional ‘exit pupil,’ which in conjunction with narrow parallel pixel beams support a high-quality image not affected by changes in eye focus. Experiments with more than 50 participants found the 3D effect to be "very convincing" for objects ranging from 20 centimeters to 10 meters off, with images in "vivid color" and high contrast.
|
Computing 'Boot Camp' is Showing 'Anyone Can Be a Coder'
Financial Times William Wallis September 10, 2019
School of Code is a U.K.-based startup that offers 15 weeks of free training at a "boot camp" for computer programming. Chris Meah set up the program to break down stereotypes about computer "nerds" and prove that "anyone can be a coder." Another of Meah's goals for the program is to address the lack of diversity in the U.K. tech industry, which is currently 81% male, according to a recent survey by business network Tech Nation. School of Code's business model relies on charging companies that employ his trained coders a recruitment fee, as well as various sponsorships. In its first year, School of Code received 300 applicants for 19 places; it received 500 applicants for 24 places in the second year. Said Meah, "We take people who are earning very little and get them on to a basic salary of typically £27,000 (around $33,500). That can go up to £30,000 or £40,000 after a year—the demand is just so high."
|
AI Speeds Photodynamics Simulations
University of Vienna (Austria) Stephan Brodicky September 11, 2019
Researchers at the University of Vienna in Austria have developed a method of using artificial neural networks to accelerate the simulation of light-induced processes significantly. The approach provides new possibilities for a better understanding of biological processes, such as the first steps of carcinogenesis or the aging processes of matter. The team taught the system complex quantum-mechanical relationships by performing some calculations beforehand and passing that knowledge on to the neural network. As part of the study, the researchers completed photodynamics simulations of a test molecule called methylenimmonium cation. After two months of computing, the researchers were able to reproduce the reaction for a nanosecond; using previous methods, the simulation would have taken about 19 years.
|
|