Data Science newsletter – December 10, 2020

GROUP CURATION: N/A

It offers “reactive embedded intelligence” bringing #AI to battery-free sensor systems and you can see it right here

A new machine learning approach offers important insights into catalysis, a fundamental process that makes it possible to reduce the emission of toxic exhaust gases or produce essential materials like fabric.

In a report published in Nature Communications, Hongliang Xin, associate professor of chemical engineering at Virginia Tech, and his team of researchers developed a Bayesian learning model of chemisorption, or Bayeschem for short, aiming to use artificial intelligence to unlock the nature of chemical bonding at catalyst surfaces.

“It all comes down to how catalysts bind with molecules,” said Xin. “The interaction has to be strong enough to break some chemical bonds at reasonably low temperatures, but not too strong that catalysts would be poisoned by reaction intermediates. This rule is known as the Sabatier principle in catalysis.”

This report, updating the September 2020 Impact Report on COVID-19, Jails, and Public Safety, draws on a sample of approximately 19 million daily jail records collected by New York University’s Public Safety Lab between Jan. 1 and Oct. 22, 2020. It explores how bookings, releases, and rebooking rates changed during the pandemic, relative to the pre-pandemic period.

In October 2013, I attended the launch of the Human Brain Project in Lausanne, Switzerland, as correspondent for Nature. I hoped to leave with a better understanding of the exact mission of the baffling billion-euro enterprise, but I was frustrated. Things became clear the following year, when the project fell spectacularly, and very publicly, apart.

Noah Hutton’s documentary In Silico captures a sense of what it was like behind the scenes of the project, which was supported with great fanfare by the European Commission. It had been hyped as a quantum leap in understanding how the human brain works. Instead, it left a trail of angry neuroscientists across Europe. Yet aspects of what went so expensively wrong still remain elusive.

Addressing inequities in access to federal and state benefit programs for people with disabilities.

Locating an immediate relief to rising CO2 emissions using fossil fuels.

Studying the geometry of neural networks to detect bias in artificial intelligence.

These are among 41 multidisciplinary endeavors that have been selected to receive support this year from Johns Hopkins University’s Discovery Awards program. Each project team is made up of members from at least two JHU entities who aim to solve a complex problem and expand the horizons of knowledge.

The University of Florida has awarded 20 faculty teams $50,000 each from UF Research’s Artificial Intelligence Research Catalyst Fund to pursue a wide range of AI-related projects. The researchers will utilize the university’s world-leading computing capabilities to analyze vast amounts of data and predict solutions to health, agriculture, engineering and educational challenges.

A team of faculty reviewers evaluated 133 proposals from across the university before settling on 20 that the group determined had the most potential for elevating UF’s AI research profile.

The projects will leverage the university’s new computing capabilities, which are being developed through $60 million in gifts from alumnus Chris Malachowsky and NVIDIA, the company he co-founded.

Thousands of college students have returned to the Boston area after the Thanksgiving break and they pose a risk of carrying COVID-19 infections with them. But researchers say colleges in Massachusetts and the rest of New England so far have been national models in testing frequently and containing the virus.

“New England is crushing it,” says Chris Marsicano, who directs the College Crisis Initiative at Davidson College, which has been tracking colleges’ COVID plans.

Marsicano’s research team finds just eight percent of colleges nationwide required testing as students left for Thanksgiving.

“It’s incredibly concerning,” Marsicano said. “Whenever students move from one point to another, they spread the disease with them.”

In New England, though, 90 percent of colleges offering some in-person classes did frequent testing before the break. “There’s no region that has a greater proportion of its institutions testing weekly or bi-weekly than New England,” Marsicano said.

A promising testing method Duke University fine-tuned this fall is one example schools can follow this spring — scanning miniature pools of mixed student samples to quickly look for signs of disease.

But experts warn the effort took extensive planning, major campus buy-in, money and help from relatively low community infection rates. Without that kind of surveillance, colleges risk becoming hot spots where the virus races through dorms, parties and surrounding areas.

“They’re not ready, by and large,” said Chris Marsicano, director of the College Crisis Initiative at Davidson College.

Early projections from Marsicano’s team say about 60 percent of U.S. higher education institutions plan to host classes with all or some portion of their students on campus in 2021. Only an estimated 8 percent of them are prepared to test each of their students at least once a week.

A sweeping new review of national test data suggests the pandemic-driven jump to online learning has had little impact on children’s reading growth and has only somewhat slowed gains in math. That positive news comes from the testing nonprofit NWEA and covers nearly 4.4 million U.S. students in grades three through eight. But the report also includes a worrying caveat: Many of the nation’s most vulnerable students are missing from the data.

“Preliminary fall data suggests that, on average, students are faring better than we had feared,” says Beth Tarasawa, head of research at NWEA, in a news release accompanying the report.

“While there’s some good news here, we want to stress that not all students are represented in the data, especially from our most marginalized communities.”

Rural Living: World-class people will move to smaller cities, have a lower cost of living & higher quality of life

These regions must innovate quickly to attract that wealth. Better schools, faster internet connections are a must

Alarm clock
Asynchronous Work: Offices are instantaneous gratification distraction factories where synchronous work makes it impossible to get stuff done

Tools that enable asynchronous work are the most important thing globally remote teams need. A lot of startups will try to tackle this

The conference on Neural Information Processing Systems (NeurIPS) 2020 kicked off on Sunday 6th December and will run until Saturday 12th December. Here, we give a brief summary of many of the planned sessions and events for the week ahead.

Surfers know they share the waters they love with sharks, but technology may soon offer some added protection from a possible encounter.

According to Southern Cross University researcher Andrew Colefax, the day is nearing that autonomous drones — which do not require a line-of-sight operator — will be able to offer shark detection at any point along the coastline.

He has spent four years of intense research and development in the field of drones and shark detection, and said artificial intelligence (AI) and machine learning will be a game changer on beaches in the near future.

“We had this idea that computational methods could estimate what the sample would look like without actually killing the cells,” said Mikhail Kandel, a graduate student in the Popescu group.

The researchers first imaged the cells over several days using their non-destructive label-free technique. At the end of the experiment, they stained the samples and used deep learning, which is a subset of machine learning, to learn where the fluorescence dyes would be located. “This let us estimate the stain in our initial movies without actually staining the cells,” Kandel said.

“Although AI has been used in the past to create one type of imaging from a different type of staining, we were able to program it to analyze the images in real time,” Popescu said. “Using deep learning, we were able to look at cells that had never been tagged with any dye, and the algorithm was able to precisely locate different parts of the cell.”

The University of Florida began construction on the Malachowsky Hall for Data Science & Information Technology, a 263,000-square-foot academic building located in the heart of UF’s main campus that will connect students and researchers from across disciplines and create a hub for advances in computing, communication and cyber-technologies with the potential for profound societal impact.

The building, anchored by a gift from UF alumnus Chris Malachowsky as well as funding from the state, will provide collaboration space and will focus on the application of computing, communication, and cyber technologies to a broad spectrum of areas including health care, pharmacology, security, technology development, and fundamental science. Malachowsky, alongside Silicon Valley computer company NVIDIA, which he co-founded, is a key partner for UF in the artificial intelligence space, including the university’s initiative to integrate AI across curriculum. The building is slated for completion around April 2023.

Mo Ebrahimkhani partnered with Patrick Cahan, Ph.D., at Johns Hopkins University to use a machine-learning system that can reverse engineer the genes necessary for human liver maturation.

Then, Ebrahimkhani together with his collaborator at Pitt, Samira Kiani, M.D., applied genetic engineering techniques, including CRISPR, to turn a mass of immature liver tissue–originally derived from human stem cells–into what the team calls “designer liver organoids.”

The more mature the organoids got, the more capillaries and rudimentary bile duct cells snaked their way through the thin sheet of tissue, and the more closely the function of the tiny organ rivaled its full-size natural human model. Energy storage, fat accumulation, chemical transport, enzyme activity and protein production were all closer to adult human liver function, though still not a perfect match.

I proposed what would become Lacework in the Summer of 2019. In my proposal, I describe a cycle of videos curated from MIT’s ‘Moments In Time’ dataset, each then slowed down, interpolated, and upscaled immensely into imagined detail, one flowing into another like a river. … When I first started watching the dataset I assumed that the team of researchers who had put it together at MIT had seen the bulk of it, but I’m now convinced that assumption was wrong. This is because so much of the archive is so, so hard to watch.

1. Start by acknowledging the person’s individual fears & concerns about vaccines. Ultimately, many of these are things we share–otherwise we wouldn’t do clinical trials to assess safety and effectiveness.

2. Acknowledge a common goal too. If the reason they don’t want a vaccine is that they’re worried about their health outcomes or their child’s health outcomes–ding ding ding, that’s common ground. You also want good health outcomes for them!