Engineering researchers have created ultrathin, stretchable electronic material that is gas permeable, allowing the material to “breathe.” The material was designed specifically for use in biomedical or wearable technologies, since the gas permeability allows sweat and volatile organic compounds to evaporate away from the skin, making it more comfortable for users – especially for long-term wear.
“The gas permeability is the big advance over earlier stretchable electronics,” says Yong Zhu, co-corresponding author of a paper on the work and a professor of mechanical and aerospace engineering at North Carolina State University. “But the method we used for creating the material is also important because it’s a simple process that would be easy to scale up.”
Specifically, the researchers used a technique called the breath figure method to create a stretchable polymer film featuring an even distribution of holes. The film is coated by dipping it in a solution that contains silver nanowires. The researchers then heat-press the material to seal the nanowires in place.
Motivated by the increasing demand for skilled data scientists, the University of San Diego has launched an innovative online degree program – the Master of Science in Applied Data Science (MS-ADS).
“The goal of the USD MS-ADS online program is to prepare prospective data scientists with in-depth analytical knowledge tied to real-world applications, relevant programming ability, and critical professional skills to both succeed in their career and to create meaningful social impact,” said Ebrahim K. Tarshizi, PhD, Program Coordinator of the Applied Data Science program.
Self-supervised learning could lead to the creation of AI that’s more humanlike in its reasoning, according to Turing Award winners Yoshua Bengio and Yann LeCun. Bengio, director at the Montreal Institute for Learning Algorithms, and LeCun, Facebook VP and chief AI scientist, spoke candidly about this and other research trends during a session at the International Conference on Learning Representation (ICLR) 2020, which took place online.
A typical Ph.D. at Carnegie Mellon can take around six or seven years to achieve. That’s plenty of time for students to get to grips with the lab’s philosophy and approach to technology. FIGLAB has access to the latest components, often long before they’re accessible to most people. But their approach to these can be dazzlingly subversive: Sure, you created this expensive component to do X, but we’re going to make it do Y because, reasons.
“It often happens where we’re playing with things and we find entirely new ways to leverage them,” [Chris] Harrison said. “We might get some crazy new sensor that might be for sensing, you know, temperature inside of a steel furnace. We’re like, ‘well, what happens if you flip it upside down and put it in a smartwatch?’ Well, oh my gosh, now you can do authentication based on blood vessels.”
After analyzing 95 million traffic stop records, filed by officers with 21 state patrol agencies and 35 municipal police forces from 2011 to 2018, researchers concluded that “police stops and search decisions suffer from persistent racial bias.”
Harvard Business Review; James DeVaney, Gideon Shimshon, Matthew Rascoff and Jeff Maggioncalda
from
As the emergency subsides but normal fails to return, higher ed institutions need to do more. There’s a good likelihood that virtual learning — in some capacity — will need to be a part of education for the foreseeable future. Higher ed institutions need a response framework that looks beyond the immediate actions. They have to prepare for an intermediate period of transition and begin future-proofing their institutions for the long term.
Building Mature Digital Learning Ecosystems
Evolution in the higher education ecosystem happens through “punctuated equilibrium”: long periods of relatively slow change interspersed with occasional moments of rapid adaptation. The current pandemic is a punctuation moment. Educators, faced with unprecedented urgency, are working hard to restore teaching and learning using technology, innovation, and collaboration.
So is dirty air, which already kills at least 7 million people a year, turbo-charging the coronavirus pandemic?
The overlap of highly polluted places, such as northern Italy, and pandemic hotspots is stark and preliminary studies point in this direction, while a link between the 2003 Sars outbreak and dirty air is already known.
Confirming the impact of air pollution on the severity of Covid-19 could make a real difference by showing the response should be boosted in places with poor air quality. But doing the scientific studies required in the midst of a global pandemic and with imperfect data is difficult.
“We had to be prepared for scale here. That isn’t just the app and the app store. The digital side is very easy to scale, but you had to be prepared to handle a massive intake of study visits and demand,” he said. “What if a whole bunch of people had AFib all along and day one of enrollment all got the notification and all called American Well? So, we had to build for exceptions and scale, which we did.”
While researchers were prepared to scale, there were also several lessons to be learned throughout the process that the partners are still addressing. In the study, around 2,200 participants got notifications about possible AFib. However, only 945 of those participants followed up with a study visit.
“On one hand, that is great. We had almost 1,000 study visits. On the other hand, we had significant attrition and loss of engagement,” Turakhia said.” So why was that? What can we improve? How can we make this more frictionless and study participants more adherent to procedures? It’s not like we have a study nurse at a local hospital that you know, and have a relationship with – and can call them. So, there is a lot here to learn.”
In the first half of 2020 alone, TV has embraced A.I. in a big way, bringing big ideas to the small screen. But they’re almost entirely dystopian. Devs, Altered Carbon, Star Trek: Picard, Ghost in the Shell: SAC_2045, and Westworld all tackle artificial intelligence (with varying degrees of intensity, philosophy, and execution); major spoilers for all of those series follow below. Some of these stories posit a highly advanced A.I. that’s then put to use on a ridiculous task, others see their machine creations committing the same sins as man; all of them show a disturbingly dark turn toward the dystopian and nihilistic. Honestly, it’s all been done before. So when the gunfire quiets and the philosophy fades, all we’re left with is a familiar story of the dangers we pose to each other, told through the lens of artificially intelligent beings. What we need, however, is something more positive, stories that inspire and encourage and explore.
What we need in the television landscape is an A.I. revolution.
Pew Research Center, FactTank; Mark Hugo Lopez, Lee Rainie and Abby Budiman
from
The coronavirus outbreak has altered life in the United States in many ways, but in key respects it has affected black and Hispanic Americans more than others.
The financial shocks of the outbreak have hit Hispanic and black Americans especially hard. When it comes to public health, black Americans appear to account for a larger share of COVID-19 hospitalizations nationally than their share of the population. And in New York City, death rates per 100,000 people are highest among blacks and Hispanics.
Safe Blues works by sending out periodic pings from smartphones equipped with Bluetooth. Through the strength of the Bluetooth signal, the tool can assess the proximity of other phones, as well as how long they remain near each other. One phone, carrying several strains of the Safe Blues virus, then communicates with a nearby phone and randomly transmits a strain of the virus – just as a human host might.
The data is sent to a central server, which then uses statistical machine learning and artificial intelligence models to predict how COVID-19 will spread.
“A prospective employee can look at all of the cultural dimensions in the heat map and see that the people in a particular unit or division of a company speak favorably about competitiveness and then say, ‘you know what, I like a competitive atmosphere, maybe I’ll self-select myself into this opportunity,” said Das Swain.
To create the 41 dimensions of organizational culture that serve as the framework for their research, Das Swain and Saha started with publicly available job descriptors from Occupational Information Network (O*NET), an occupational information database sponsored by the U.S. Department of Labor. The job descriptor language is rooted in occupational psychology. It also reflects current workplace norms and expectations for nearly 200 professional and technical occupations.
With this database, the team developed 41 computer-readable language models. Models like these are known as word vectors and can reveal often-unseen connections between seemingly unrelated words, compound words, phrases, as well as word parts like prefixes and suffixes.
McKinsey; Hamilton Boggs, Charag Krishnan, Samvitha Ram, and Jimmy Sarakatsannis
from
When it comes to enrollment leading up to the start of classes each fall, enrollment management teams and admissions offices across US universities are accustomed to facing uncertainty. The COVID-19 pandemic has exacerbated this phenomenon. Typically, universities ramp up their high-contact admissions activities in March and April; this year they have faced unprecedented uncertainty around virtual engagement models and enrollment figures. Indeed, a quarter of students who had decided on a school are reconsidering where to enroll, and 20 percent say it’s likely they won’t attend a higher education institution in the fall. 1 In a recent survey that asked university presidents to name their primary worries about the pandemic, 42 percent had enrollment at the top of their lists. 2
These uncertainties come on top of recent changes to the National Association for College Admission Counseling (NACAC) Code of Ethics and Professional Practice that have upended how universities think about the admissions cycle. Historically, universities could rest assured that the number of students who had deposited with them by May 1 served as a reasonable indicator of fall enrollment.
Residents of both places typically have household income below the Queens median and a similar share of people who lack health insurance, as measured by the U.S. Census Bureau. And almost half of apartments and houses in both areas have more than one occupant per room, the Census definition of crowded.
Yet when it comes to COVID-19, the differences between the neighborhoods couldn’t be more stark.
Corona emerged as the early epicenter of the outbreak in New York City and shows no sign of slowing down. Meanwhile, the rate of test-confirmed positive cases of the virus among Flushing residents has remained among the lowest in the five boroughs.
A group of biodiversity experts warned that future pandemics are on the horizon if mankind does not stop its rapid destruction of nature.
Writing an article published Monday by The Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES), the authors put the responsibility for COVID-19 squarely on our shoulders.
The growing COVID-19 pandemic could last up to two years, with a potential second wave in the fall, according to nationally renowned University of Minnesota epidemiologist Michael Osterholm and a team of researchers.
The exact path the disease will take is unclear, but with no vaccine and a global population that had no immunity to the new coronavirus, COVID-19 could follow patterns seen in previous pandemics.
That means that governments will need to continually adjust their pandemic responses to waves of infections, which could have several peaks, rather than a distinct period of illnesses that burns out in a matter of months.
“We are learning how to die with this virus and we have to learn how to live with it,” said Osterholm, director of the Center for Infectious Disease Research and Policy at the U and a frequent infectious disease expert on national cable news programs.
A team of scientists has taken steps to create a new form of digital data storage, a “Racetrack Memory,” which opens the possibility to both bolster computer power and lead to the creation of smaller, faster, and more energy efficient computer memory technologies.
“Racetrack memory, which reconfigures magnetic fields in innovative ways, could supplant current methods of mass data storage, such as flash memory and disk drives, due to its improved density of information storage, faster operation, and lower energy use,” says Yassine Quessab, a postdoctoral fellow at New York University’s Center for Quantum Phenomena (CQP) and the lead author of the work, which is reported in the journal Scientific Reports.
“While additional development is necessary in order to deploy them in consumer electronics, this pioneering type of memory may soon become the next wave of mass data storage,” adds NYU Physics Professor Andrew Kent, the paper’s senior author.
“The Postdoctoral Fellows leverage NEON data in collaboration with NEON staff and community contributors to generate scientific outcomes and peer-reviewed publications, while engaging the scientific community in NEON data use and potentially generating new tools to facilitate use of NEON data.” Deadline for applications is May 8.
Santa Barbara, CA September 21-22. “Explore future directions and opportunities for synthesis in ecology and environmental science and help set a research agenda for synthesis ecology in the next decade.” Deadline for applications is May 15.
“The coronavirus outbreak in China and around the globe poses challenges for the economies, foreign policies, and societies of China and the world at large, as well as public-health concerns for those infected or at risk. Help to forecast how this disease will shape the world order.” Tournament closes on January 1, 2021.
This paper introduces GIMP-ML, a set of Python plugins for the widely popular GNU Image Manipulation Program (GIMP). It enables the use of recent advances in computer vision to the conventional image editing pipeline in an open-source setting. Applications from deep learning such as monocular depth estimation, semantic segmentation, mask generative adversarial networks, image super-resolution, de-noising and coloring have been incorporated with GIMP through Python-based plugins. Additionally, operations on images such as edge detection and color clustering have also been added. GIMP-ML relies on standard Python packages such as numpy, scikit-image, pillow, pytorch, open-cv, scipy. Apart from these, several image manipulation techniques using these plugins have been compiled and demonstrated in the YouTube playlist (this https URL) with the objective of demonstrating the use-cases for machine learning based image modification. In addition, GIMP-ML also aims to bring the benefits of using deep learning networks used for computer vision tasks to routine image processing workflows. The code and installation procedure for configuring these plugins is available at this https URL.
“In this article, we talk about Bayesian Optimization, a suite of techniques often used to tune hyperparameters. More generally, Bayesian Optimization can be used to optimize any black-box function.”
This post outlines how to integrate a state-of-the-art machine learning model into a Slack bot to generate a summary of articles shared by their URL — a.k.a “tldr” (too long, didn’t read). To round up the rest of the tech stack, I’ll cover how to set up Cortex (Machine learning model deployment platform), Serverless Framework (AWS Lambda), and AWS Simple Queue Service (SQS).