Data Science newsletter – April 17, 2021

Newsletter features journalism, research papers and tools/software for April 17, 2021

 

Twitter will share how race and politics shape its algorithms

Protocol, Anna Kramer


from

Twitter is studying the biases within its algorithms as part of a new effort to try to understand how its machine learning tools can cause unintended consequences, and the company says it plans to publicly share some of its findings.

The company plans to prioritize what it’s calling the pillars of “responsible ML,” which include “taking responsibility for our algorithmic decisions, equity and fairness of outcomes, transparency about our decisions and how we arrived at them, and enabling agency and algorithmic choice,” according to a Wednesday announcement outlining this approach. Rumman Chowdhury, the leader of Twitter’s ML Ethics, Transparency and Accountability team, co-wrote the statement with Jutta Williams.


University at Albany’s New 1.8 Megawatt Solar System Supports ‘Energy Master Plan’

Environment + Energy Leader, Jennifer Nastu


from

The University at Albany-State University of New York (SUNY) has begun construction on a solar project that will supply about 60% of the electricity used by its new $180 million ETEC building, expected to open this summer. The university will purchase all the electricity and renewable energy credits (RECs) produced through the 1.8 megawatt system. The new system — one of the many projects in the university’s energy master plan — is being constructed on the buildings of the university’s “Academic Podium,” the buildings surrounding its open courtyard.

In the first year, it will offset more than 3.6 million pounds of carbon dioxide; it will also provide long-term financial savings and help the ETEC building achieve LEED Platinum certification, UAlbany says. The university hopes to reduce its carbon footprint by 40% by 2030.


Questioning the role of AI in exam marking

Raconteur (UK), Sam Haddad


from

Dee Kanejiya, founder and chief executive of Cognii, an AI-based platform that uses natural language conversion to assess passages of longer text that have traditionally been harder for AI to grade accurately, wants to help correct what he sees as an over-reliance on multiple choice questions in US assessments.

He believes these do not help students in the real world and is a format that favours boys over girls. But marking longer answers is time consuming for teachers and therefore expensive, which is where he hopes Cognii can help.

Kanejiya is excited about the potential of AI to free up teachers from repetitive tasks such as marking, though he insists it isn’t about replacing them. “You get more time for that intimate relationship between faculty and students if teachers are not grading,” he says. “They can spend more quality time with the students, time for the emotional side of things, which they’re good at.”


Higher Education’s Increasingly Nuanced, AI-Powered Chatbots

EdTech Magazine, Erin Brereton


from

Years ago, any questions students had about issues such as enrollment, academics or housing could be asked and answered only during designated hours, whether in person or by phone.

Today, however, a number of schools are turning to artificial intelligence–powered chatbots that crank out detailed responses 24 hours a day, using data pulled from a knowledge base that acts as a central repository of information, says Patricia Velazquez, director of education and research industry strategy at Oracle, which offers a digital assistant that is used in higher education environments.

“COVID definitely highlighted the need for digital assistants,” Velazquez says. “Phone lines were overloaded in the beginning of the pandemic. Users were unsure where they’d get information. A digital assistant really helps in an instance like this when it can provide accurate and to up-to-date information across the institution by pulling it from a single source.”


University of Iowa administrators explore plans to establish a new student Well-Being Center

The Daily Iowan student newspaper, Brady Osborne and Marco Oceguera


from

Students may eventually see a new Well-Being Center on campus that would cohesively combine various student health services into one building.

The Division of Student Life, in partnership with Undergraduate Student Government and Graduate and Professional Student Government, recently announced preliminary plans for the project that would construct the center, which is to be entirely funded by a new student activity fee. The dollar amount for the added fee won’t be determined until construction plans and a final price tag is finalized.

The push for a new Well-Being Center is a response to a recent National College Health Assessment Survey, which collects data on student health interests and behaviors each year. The data highlighted recent declines in students’ physical and mental health. Administrators have also cited the deteriorating conditions of Westlawn Hall, where the current Student Health Center is located.


16 states back Alabama’s challenge to Census privacy tool

Herald Mail Media, Associated Press, Mike Schneider


from

Sixteen other states are backing Alabama’s challenge to a statistical method the U.S. Census Bureau is using for the first time to protect the privacy of people who participated in the 2020 census, the nation’s once-a-decade head count that determines political power and funding.

A federal judge on Monday allowed the 16 states to file a brief in a support of a lawsuit brought by Alabama last month. The suit seeks to stop the Census Bureau from applying the method known as “differential privacy” to the numbers that will be used for redrawing congressional and legislative seats later this year.

The states supporting Alabama’s challenge are Alaska, Arkansas, Florida, Kentucky, Louisiana, Maine, Mississippi, Montana, Nebraska, New Mexico, Ohio, Oklahoma, South Carolina, Texas, Utah and West Virginia. Maine and New Mexico have Democratic attorneys general, while all the other states have Republican ones.


“Genius Makers”: A Conversation w/ Cade Metz & Yann LeCun

Medium, NYU Center for Data Science


from

On Friday, April 9, CDS sat down with New York Times reporter Cade Metz and CDS Founder and Professor of Computer Science, Neural Science, Data Science, and Electrical and Computer Engineering Yann LeCun to discuss Metz’ new book, Genius Makers: The Mavericks Who Brought AI to Google, Facebook, and the World. Hosted by CDS Director Julia Kempe, and moderated by CDS & Courant CS Research Professor Michael Picheny, the conversation was joined by over 100 attendees, both inside and outside the NYU community.

Picheny centered the lively discussion around four key areas: (1) Retrospective Questions about AI Development, (2) the Current Status of AI, (3) Social issues that have been created by the recent success of AI, and (4) Future Directions. Questions probed the origin of Metz’ book and LeCun’s interest in AI, as well as LeCun’s reaction to being featured prominently in Metz’ book (“I don’t see anything wrong in the book — which is great, it’s a great compliment coming from me! I think it puts a searchlight on one aspect which is … industry oriented a little bit, and there’s another aspect which is more the research.”)


Google is poisoning its reputation with AI researchers

The Verge, James Vincent


from

“Not only does it make me deeply question the commitment to ethics and diversity inside the company,” Scott Niekum, an assistant professor at the University of Texas at Austin who works on robotics and machine learning, told The Verge. “But it worries me that they’ve shown a willingness to suppress science that doesn’t align with their business interests.

“It definitely hurts their credibility in the fairness and AI ethics space,” says Deb Raji, a fellow at the Mozilla Foundation who works on AI accountability. “I don’t think the machine learning community has been very open about conflicts of interest due to industry participation in research.”


VW Comes To Detroit To Develop Its Artificial Intelligence

Carbuzz, Karl Furlong


from

The carmaker has an entire team dedicated to automotive AI tech in the Motor City.

We’re only starting to get to grips with the benefits of artificial intelligence in modern vehicles now. Recently, Hyundai announced new AI-based voice recognition technology for the 2021 Elantra that can learn the driver’s habits, improving interaction with the system. Chrysler and Jeep vehicles are set to use a similar system called Alexa Custom Assistant.

But AI’s benefits extend well beyond the vehicle and the driver; the tech also has advantages in automotive factories and on the business side as well. Volkswagen is digging deeper into this with its new AI Detroit unit, a research and development unit with its home in the Motor City.


Biosensor to protect chocolate supplies

ee News Europe, Nick Flaherty


from

Most chocolate consumers are naturally unaware of infections that can damage cocoa production; the cacao swollen shoot virus (CSSV) infection can lead to the death of cocoa trees and threaten the livelihoods of cocoa farmers.

Researchers at the University of the West of England (UWE) are developing a handheld device which identifies whether or not the plant has the virus before any symptoms appear. The timely detection of disease enables early intervention to prevent it from spreading further, as well as offering economic benefits.


After Cookies, Ad Tech Wants to Use Your Email to Track You Everywhere

Electronic Frontier Foundation, Bennett Cyphers


from

There are several proposals from ad tech providers to preserve “addressable media” (read: individualized surveillance advertising) after cookies die off. We’ll focus on just one: Unified Identifier 2.0, or UID2 for short, developed by independent ad tech company The Trade Desk. UID2 is a successor to The Trade Desk’s cookie-based “unified ID.” Much like FLoC, UID2 is not a drop-in replacement for cookies, but aims to replace some of their functionality. It won’t replicate all of the privacy problems of third-party cookies, but it will create new ones.

There are key differences between UID2 and Google’s proposals. FLoC will not allow third-party trackers to identify specific people on its own. There are still big problems with FLoC: it continues to enable auxiliary harms of targeted ads, like discrimination, and it bolsters other methods of tracking, like fingerprinting. But FLoC’s designers intend to move towards a world with less individualized third-party tracking. FLoC is a misguided effort with some laudable goals.

In contrast, UID2 is supposed to make it easier for trackers to identify people.


FSF doubles down on Richard Stallman’s return: Sure, he is ‘troubling for some’ but we need him, says org • The Register

The Register, Thomas Claburn


from

The Free Software Foundation (FSF) on Monday apologized for mishandling the announcement last month that founder Richard Stallman, or RMS, had been reelected to its board of directors – and published a statement from RMS both justifying his behavior and apologizing for it.

“FSF staff should have been informed and consulted first,” the FSF said. “The announcement by RMS at LibrePlanet was a complete surprise to staff, all those who worked so hard to organize a great event, to LibrePlanet speakers and to the exhibitors. We had hoped for a more inclusive and thoughtful process and we apologize that this did not occur.”


The SAME data were given to 73 research teams to test the SAME hypotheses on immigration and policy support.

Twitter, Scott Page


from

The result: 17% support, 58% no effect, 25% reject. Evidence that empiricists rely on different models to organize data.


Colorization APIs are becoming widespread; AI-colorized historical photos are circulated without caveat. But is AI colorization providing an accurate image of the past? To find out, I digitally desaturated these color photos by Sergey Prokudin-Gorsky

Twitter, Gwen C. Katz


from

I then colorized the photos using the DeepAI Image Colorization API: https://deepai.org/machine-learning-model/colorizer

Here’s the AI’s attempt to color these photos.

And here are the original color photos.


Overcoming financial limitations in global volcano monitoring

Nature Communications, Opinion


from

Equitable partnerships among the international volcano science community are important now more than ever, to cope with financial disparities and ultimately allow for worldwide volcano monitoring oriented to hazard mitigation.

SPONSORED CONTENT

Assets  




The eScience Institute’s Data Science for Social Good program is now accepting applications for student fellows and project leads for the 2021 summer session. Fellows will work with academic researchers, data scientists and public stakeholder groups on data-intensive research projects that will leverage data science approaches to address societal challenges in areas such as public policy, environmental impacts and more. Student applications due 2/15 – learn more and apply here. DSSG is also soliciting project proposals from academic researchers, public agencies, nonprofit entities and industry who are looking for an opportunity to work closely with data science professionals and students on focused, collaborative projects to make better use of their data. Proposal submissions are due 2/22.

 


Tools & Resources



James Landay: Smart Interfaces for Human-Centered AI

YouTube, Stanford HAI


from

AI has the potential to automate people out of their jobs, and in some cases, it will. But while we should carefully consider the risk of replacing human capabilities, it’s important to realize that AI has enormous potential to augment them as well: it can boost the creativity of our work, help us learn better, deliver healthcare more effectively, and make our societies more sustainable. Like any tool, however, AI and its relationship with humans has as much to do with its interface as it does with the underlying capabilities it provides. Does it amplify our actions and remain attentive to our goals—even as we revise them—or is it a black box that accomplishes tasks autonomously? [video, 59:40]


How junior scientists can land a seat at the leadership table

Nature, Career Feature, Kendall Powell


from

Steps by junior researchers to claim seats at science’s decision-making tables are becoming more common. Doctoral students, postdocs and people who have had their PhD for less than ten years are joining advisory boards, oversight councils and conference-organizing committees. Others have started their own advocacy and research initiatives by founding non-profit organizations and companies, bringing fresh perspectives and up-to-date expertise to boardrooms and advisory committees. And they gain organizational, management and leadership experience.


Telling a Great Data Story: A Visualization Decision Tree

KDnuggets, Stan Pugsley


from


The @scikit_learn team at @sklearn_inria has created a MOOC for getting started with scikit-learn! Go check it out!

Twitter, Andreas Meuller


from

https://inria.github.io/scikit-learn-mooc/

You can leave feedback and suggestions (or make PRs) on the repo here


Careers


Internships and other temporary positions

Technology / Science Writer



Good AI; Prague, Czech Republic, or Remote
Full-time, non-tenured academic positions

Full-time Research Assistant Position



Northeastern University, Network Science Institute; Boston, MA

Leave a Comment

Your email address will not be published.