Northeastern University Institute for Experiential AI
Welcome to the October 23, 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."

Students in the Nairobi Play Project, which teaches computing to groups in or at risk of conflict CS Classes Can Break Down Cultural Barriers
Cornell Chronicle (NY)
Melanie Lefkowitz
October 21, 2019

A study found that computer science (CS) classes can surmount cultural barriers and defuse conflict among groups of culturally dissimilar people. Researchers from Cornell Tech, the University of Michigan, and the Nairobi Play Project focused on the Kenyan initiative, a United Nations-funded program designed to foster intercultural learning between groups in or at risk of conflict. The project held 30 after-school sessions in a Kenyan refugee camp, led by educators who were themselves refugees, to teach students computing fundamentals and have them collaboratively create video games with community-based themes. The researchers determined the CS classes' structure nurtured unlikely friendships between students from disparate backgrounds. The limited number of devices in class necessitated cooperation between students, while the need for coders to consider problems from the computer's perspective also helped cultivate intercultural connections.

Full Article

Google’s Quantum computer Google Claims Quantum Breakthrough That Could Change Computing
The New York Times
Cade Metz
October 23, 2019

Google said it has achieved a breakthrough called quantum supremacy, which could speed computational calculations far beyond the capabilities of classical systems. The search giant said its quantum computer completed a mathematical calculation in less than four minutes that the largest supercomputers could not complete in less than 10,000 years. IBM disputed Google's statement, arguing that a current computer system could theoretically perform the calculation in less than two and a half days. Some researchers have dismissed Google's achievement for the esoteric nature of the calculation, while Chad Rigetti, who worked on IBM's quantum computing project and now runs his own start-up, hailed it as a major achievement for the broader scientific community. Said Rigetti, “It is just a short amount of time now before we have commercially relevant problems that quantum machines can solve.”

Full Article
*May Require Paid Registration
UTSA Study Warns of Security Gaps in Smart Bulbs
University of Texas at San Antonio
Milady Nazir
October 22, 2019

Researchers at the University of Texas at San Antonio (UTSA) have examined security holes that exist in popular smart-light brands, and found many of them could be prime targets for hackers. Some smart bulbs connect to a home network without needed a smart home hub. If these same bulbs are also infrared-enabled, hackers can send commands via the infrared invisible light emanating from the bulbs to either steal data or spoof other connected Internet of Things devices on the home network. Said UTSA's Murtuza Jadliwala, "These bulbs are now poised to become a much more attractive target for exploitation even though they have very simple chips."

Full Article

A doctor listening to the sound of a baby’s heartbeat Is the Stethoscope Dying? High-Tech Rivals Pose Threat
Associated Press
Lindsey Tanner
October 23, 2019

New medical technologies competing with the two-century-old stethoscope include handheld devices that use ultrasound, artificial intelligence (AI), and smartphone apps. Advocates say these devices are as easy to use as stethoscopes, but allow doctors to monitor the body in motion and to visualize leaky valves and other problems. Medical students at Indiana University continue to learn stethoscope skills, but also receive training in hand-held ultrasound in a program launched last year by Dr. Paul Wallach. Wallach created a similar program five years ago at the Medical College of Georgia, and predicts hand-held ultrasound devices will become part of a routine physical exam within a decade.

Full Article
Researchers Identify Cancer Symptom Clusters Using EHR Data
Kate Anderton
October 22, 2019

Researchers at the Regenstrief Institute, Indiana University-Purdue University Indianapolis, and Indiana University’s schools of Medicine and Nursing have developed techniques that use data on breast and colorectal cancer symptoms from electronic health records (EHRs) to explore potential links between each disease and symptom clusters. Said Regenstrief's Kun Huang, "Identifying and understanding symptom clusters—which symptoms tend to go together—fatigue and depression, for example—and when these symptoms occur during the course of treatment—provides critical information to a patient's care team, especially as we look forward to precision health." The new techniques combine data mining, natural language processing, and machine learning to analyze clinical notes in EHRs and extract data from structured entries in medical records. Algorithms developed by the researchers determined symptom clusters differed for breast cancer and colorectal cancer patients.

Full Article
When Wi-Fi Is Weak, Send Noise Instead
The Source (Washington University in St. Louis)
Brandie Jefferson
October 22, 2019

Washington University in St. Louis researchers overcame a loss of Wi-Fi signal when gathering indoor air-quality readings from sensors by sending noise. The team programmed a series of 1s and 0s into a Wi-Fi sensor, activating and deactivating the signal in a specific pattern so the router could distinguish the signal from ambient wireless noise. The team also demonstrated that the code's transmission could be extended outside the limits of the standard Wi-Fi data range. Said researcher Philip Lundrigan, “We can send and receive data regardless of what Wi-Fi is doing. All we need is the ability to transmit energy and then receive noise measurements.”

Full Article

Brondby fans scuffle with police during a match Soccer Team In Denmark Is Using Facial Recognition to Stop Unruly Fans
Sidsel Overgaard
October 21, 2019

Brondby Stadium, on the outskirts of Copenhagen, Denmark, is using one of Europe's first large-scale private facial recognition systems to catch those who have been banned from the stadium as troublemakers. Pictures of those on a watchlist are entered into the system on game day and are deleted at the end of the day. The system is not connected to the Internet, and there is a crosscheck to avoid false positives. Said local resident Martin Lund, “I think it's great!" On the other hand Jesper Lund of watchdog group the IT-Political Association of Denmark, said, "Using this very invasive and error-prone technology for something like making sure that persons on a banned list cannot go to a football match is really not proportionate."

Full Article
Turning Chairs Into Tables: Novel Deep Learning Method for Unpaired Shape Transforms
SIGGRAPH Asia 2019
October 18, 2019

Researchers at Canada's Simon Fraser University, China's Shenzhen University, and Israel's Tel Aviv University have developed a deep neural network that can learn to perform natural unpaired shape transforms. The LOGAN network can automatically translate shapes between two domains and automatically execute content and style transfers without changing network architecture. LOGAN is trained on two sets of input shapes, and encodes them into a common latent space; it then uses a generative adversarial network technique for unpaired general-purpose shape transforms. Explained Simon Fraser’s Hao Xhang, “Shape transform is one of the most fundamental and frequently encountered problems in computer graphics and geometric modeling. What is new and emerging is to tie this important problem to deep learning—can a machine learn to transform shapes, particularly under the unsupervised or unpaired setting?”

Full Article
Browser Tool Aims to Help Researchers ID Malicious Websites, Code
NC State University News
Matt Shipman
October 22, 2019

North Carolina State University (NC State) researchers have developed a tool that can track and record the behavior of malicious JavaScript programs, without tipping off the websites that run those programs. VisibleV8 operates in the Chrome browser to identify malware that can avoid current detectors. Explains NC State's Alexandros Kapravelos, "VisibleV8 runs in the browser itself, recording how JavaScript is executed; it doesn't interact with the code and, as a result, is far more difficult to detect." VisibleV8 stores data on how a site uses JavaScript, building a behavior profile that researchers can apply to characterize malicious sites, and the utilization of JavaScript to compromise web browsers and user data. Said Kapravelos, “We’ve created a stealthy tool for monitoring JavaScript in the wild. We’re now making it open source, in hopes that it will be useful to anyone doing research on web privacy and security.”

Full Article
Haptic Arm Places Robotics Within Easy Reach
University of Bristol News
October 20, 2019

Researchers at the University of Bristol in the U.K. have created a robotic arm facilitating light, affordable, and accessible haptic force feedback. The researchers say the Mantis can be built for a twentieth of the cost of its market equivalent, by using significantly less-expensive components. Bristol's Anne Roudaut said Mantis extends humans' inherent tactile sense by allowing users to touch and feel three-dimensional objects, which will add realism to the virtual reality (VR) experience. Said Roudaut, "Humans already have a great sense of touch. Mantis expands on this innate ability by enabling people to touch and feel 3D objects, adding more depth to the VR experience.”

Full Article

A user traverses a park while wearing a DreamWalker kit Microsoft's DreamWalker VR Turns Your Daily Commute Into Totally Different One
Ars Technica
Samuel Axon
October 22, 2019

Microsoft researchers have designed virtual reality (VR) solutions to allow users to remain fully immersed in virtual environments, even when walking in real-life locations. DreamWalker involves advance planning of a real-world path through an outdoor environment and corresponding virtual setting based on geolocation data, with updates on the fly as necessary. Potential obstacles users encounter are recorded by real-time sensors in the VR equipment, including a dual-band global positioning system sensor, red-green-blue depth cameras, and a Windows Mixed Reality-provided relative position trace. Such impediments may be replaced by virtual counterparts, while arrows guide users in safe directions. DreamWalker also employs Microsoft's Mise-Unseen eye-tracking technology, so VR surroundings can be revised while the user is looking at something else.

Full Article

UPS is striking deals with medical groups to launch its commercial drone delivery services UPS Strikes Agreements to Use Drones to Deliver Medical Supplies
The Wall Street Journal
Jennifer Smith
October 21, 2019

United Parcel Service (UPS) announced agreements with healthcare groups to use drones to deliver medical samples and supplies. One agreement permits the drones to carry cargo on hospital campuses at the University of Utah and elsewhere. UPS also agreed with CVS Health to assess drone-based home delivery of prescriptions and other products. UPS also is readying a drone logistics project with managed care consortium Kaiser Permanente to evaluate building-to-building supply transfers via drone at Kaiser's 39-hospital network. UPS said medical supplies are very well-suited to drone delivery because they are lightweight and tend to be uniform in size.

Full Article
*May Require Paid Registration
Better Fact-Checking for Fake News
MIT News
Rachel Gordon
October 17, 2019

Researchers at the Massachusetts Institute of Technology's Computer Science and Artificial Intelligence Laboratory (CSAIL) have identified flaws in detectors for automatically identifying machine-generated disinformation. The team devised techniques to fool detectors, such as automatically doctoring human-written text using a GPT-2 language model during edits. The CSAIL researchers also created systems for spotting false statements, using the Fact Extraction and VERification (FEVER) fact-checking dataset. Analysis showed FEVER harbored bias that could be embedded in models trained on the dataset; the team developed a de-biased dataset to create an algorithm that outperforms earlier fact-checking programs across all metrics.

Full Article
Virginia Tech Master of Information Technology
Northeastern University Institute for Experiential AI

Association for Computing Machinery

1601 Broadway, 10th Floor
New York, NY 10019-7434

ACM Media Sales

If you are interested in advertising in ACM TechNews or other ACM publications, please contact ACM Media Sales or (212) 626-0686, or visit ACM Media for more information.

To submit feedback about ACM TechNews, contact: [email protected]