Data Science newsletter – October 24, 2017

Newsletter features journalism, research papers, events, tools/software, and jobs for October 24, 2017

GROUP CURATION: N/A

 
 
Data Science News



Machine learning online course: I just coded my first AI algorithm, and oh boy, it felt good

Quartz, Tony Chen


from

The moment I realized that I had just coded a machine to recognize a biological entity was surreal. When the full code ran without any errors, I pumped my fists and yelled out a triumphant “OH YEAH!”, startling my kids. Frankly, I was surprised at my own reaction. I may have only correctly identified a four-legged feline, but, in the moment, it felt like so much more than that.


UH Launches Institute for Data Science

University of Houston, News & Events


from

The University of Houston has launched a new Institute for Data Science, naming Andrea Prosperetti to lead the venture.

Prosperetti, a member of the National Academy of Engineering, was recruited to UH in 2016 through the Governor’s University Research Initiative, which helps fund efforts to bring nationally recognized researchers to Texas institutions of higher education.

“Houston is a natural location for a focus on data science, with its strengths in energy and health care and its substantial technology and software industries,” said Prosperetti, who also serves as director of the UH Center for Advanced Computing and Data Systems and is Distinguished Professor of Mechanical Engineering.


Get credit for your data – BMC Research Notes launches data notes

BMC Blog Network, Dirk Krüger


from

Earlier this year BMC Research Notes was re-structured with a focus on short note articles. Now we are launching data notes, short data descriptors that aim to increase data visibility and support the reuse and sharing of valuable research data.


Allen Institute for Brain Science awarded ~$100 million to lead global collaborations that will build comprehensive atlases of brain cell types

Allen Institute for Brain Science


from

The National Institutes of Health today awarded the Allen Institute for Brain Science three five-year grants totaling nearly $100 million, to expand upon work in investigating cell types in the mouse and human brains: a crucial step toward understanding brain health and disease. The grants are part of the BRAIN Initiative Cell Census Network (BICCN), which aims to establish reference cell atlases of the brain by supporting collaborations that will create resources to be shared throughout the research community. Of the nine collaborative consortia established by these BICCN grants, three will be led by Allen Institute for Brain Science researchers.


Amazon launches Handmade Gift Shop, offering more artisan items in latest blow to Etsy

GeekWire, Kurt Schlosser


from

Just in time for the holidays, Amazon launched a new gift shop for its Handmade shopping portal, offering a selection of handcrafted items in a variety of categories. It’s the latest blow to Etsy, the creative platform struggling to hold off the e-commerce giant.

Amazon Handmade originally launched in 2015, and now offers almost 90,000 products in categories including for her, for him, for kids, for baby, for the couple and more. Items range from jewelry to tie clips to cutting boards, clocks and baby bibs.


Our Federal Science Agencies Are in Mortal Danger

Scientific American Blog Network, Andrew A. Rosenberg and Kathleen Rest


from

Amid this governmental turmoil, another longer-term development is underway that will affect the lives of everyone in the United States and impact others around the world—likely for decades to come: the loss of critical expertise and capacity in the science agencies of the federal government, including agencies such as the Environmental Protection Agency, the US Fish and Wildlife Service and others in the Department of Interior. Or the Centers for Disease Control, the Occupational Safety and Health Administration (OSHA), the National Oceanic and Atmospheric Administration (NOAA), the National Aeronautics and Science Administration (NASA), among many others.


Andreessen Horowitz Leads $30M Round In Customer Data Platform ActionIQ

AdExchanger, James Hercher


from

ActionIQ on Monday raised a $30 million Series B round led by Silicon Valley VC firm Andreessen Horowitz.

New York City-based ActionIQ is a customer data platform that collects and integrates enterprise data sources.

The round brings ActionIQ’s total funding to about $45 million. The customer data platforms Segment and mParticle, which have raised $109 million and $76 million, respectively, also have announced large funding rounds in recent months.


Noelle Selin: Tracing toxins around the world

MIT News


from

In 1995, the Governing Council of the United Nations Environment Program called for a united, global effort to reduce persistent organic pollutants (POPs) — synthetic chemicals such as PCBs, DDT, and dioxins. The compounds were known to persist and accumulate far from their sources, polluting the environment and causing adverse health effects in humans.

As work on a global treaty progressed, Noelle Eckley Selin, then a college intern at the Environmental Protection Agency, had the opportunity to play a small part in the process that eventually produced the Stockholm Convention on Persistent Organic Pollutants. She was tasked in 1997 with evaluating potential chemicals as add-ons to the “dirty dozen” that the treaty proposed to regulate.

“The treaty was designed as a dynamic instrument, so countries could add chemicals to it to respond to emerging threats,” recalls Selin, now a tenured associate professor in MIT’s Institute for Data, Systems, and Society (IDSS) and the Department of Earth, Atmospheric and Planetary Sciences (EAPS). “But the EPA wanted to know what scientific criteria to use to choose those substances. So I had to go into the dusty basement of the EPA and look up how long these random chemicals persisted in the environment.”


Does faculty productivity really decline with age? New study says no

University of Colorado Boulder, CU Boulder Today


from

For 60 years, studies of everyone from psychologists to biologists to mathematicians have shown the same remarkably similar academic research trajectory: Scientists publish prolifically early in their careers, peak after about five years, get tenure and begin a long slow decline in productivity.

But a new CU Boulder study published today in the journal PNAS suggests that stereotype is misleading.

“We found that only about one-fifth of researchers have careers that actually look like that expected curve, and the other 80 percent exhibit a really diverse set of productivity trajectories,” says first author Samuel Way, a postdoctoral researcher in the Department of Computer Science.


University of Houston to start data science center

Houston Chronicle, Lindsay Ellis


from

A data center will be coming to Houston soon after all.

The University of Houston said Thursday it will start an on-campus data science center. The decision comes months after the University of Texas System called off plans to develop a 300-plus acre swath of land, likely for that purpose, in March amid a political backlash.

UH’s institute will cover cyber and physical security, drug development and discovery, sustainability and health care. Planning began 18 months ago, Provost Paula Myrick Short said in a news release. UT’s plans were made public in March.


Want to Protect Your Wall Street Job From Robots? Learn How to Code

Bloomberg, Sarah Ponczek and Dakin Campbell


from

Goldman Sachs Group Inc. realized it had a problem.

For years, the investment bank recruited top coders to help its army of traders execute strategies with software, a role that’s becoming ever more important in a new era of automation. But the technology gurus seen as the future kept quitting while traditional traders stayed.

The firm’s solution: Register coders as full traders and hand them control of their desks. That move already is shaping who will run the trading division for decades to come.


Berkeley defining the next academic frontier: Two NSF awards to UC Berkeley set the stage for a new era of Data Science

University of California-Berkeley, Division of Data Sciences


from

The Berkeley Institute for Data Science (BIDS) and its partners are pleased to announce two exciting new Data Science efforts receiving simultaneous awards from the National Science Foundation(link is external) (link is external) (NSF). Reflecting the breadth and depth of data science at UC Berkeley, the first award will deepen the theoretical foundations of data science in a new transdisciplinary institute, while the second will strengthen educational strategies through national workshops led by the faculty and staff who have guided Berkeley’s broad-ranging data science curriculum.

,br/>
The awards come as UC Berkeley moves forward in its national leadership, cemented by the growing impact of the Berkeley Institute for Data Science and the vision of integrating data science collaborations across the university with a newly created Division of Data Sciences.


Expanding computer science in schools is a bipartisan opportunity

TheHill; Alastair Fitzpayne, Matthew Sigelman and Austin Jaspers


from

A bipartisan idea is a rare creature in Washington these days, but there is one issue that brings the parties together: the need to expand computer science education in America’s schools. President Obama proposed spending an additional $4 billion, and President Trump released a more modest proposal. But despite these efforts, schools are still waiting for additional funding. That’s a shame because computer science skills hold the keys to economic opportunity for students. Just as important, the benefits don’t just accrue to students themselves. Rather, these are competencies that will be increasingly important to American competitiveness in the 21st century.

Our continued failure to fund broad access to computer science education is shortsighted and threatens to stymie our future prosperity. We live in a world defined by computer code and a job market that increasingly demands the ability to create it. Coding skills are a staple of half of all occupations in the top earning quartile, jobs with salaries of $75,000 and over. That means that computer science skills, which include coding, are increasingly a prerequisite for a chance at a middle class life in the 21st century. Yet, remarkably, only 40 percent of K-12 schools offer any classes that include coding.

 
Events



Network coding – a personal account of combining theory and practice

UMass Amherst, College of Information and Computer Sciences


from

Amherst, MA November 8. University of Massachusetts, College of Information and Computer Sciences Distinguished Lecturer Series
Speaker: Muriel Medard. [free]


Healthcare + A.I. Northwest

Xconomy


from

Seattle, WA November 9. “Xconomy is convening an elite group of healthcare and A.I. leaders from Seattle and beyond for a special half-day forum to discuss these topics, and much more.” [$$$]

 
Deadlines



Surveillance Studies Network 8th Biennial Conference

Aarhus, The Netherlands June 7-9, 2018! Deadline for proposals is December 31.
 
NYU Center for Data Science News



CDS Hosts 10th Data Science Showcase

Medium, Center for Data Science


from

Four statistics researchers (Xi Chen,
Jennifer Hill, Mengling Liu, David Madigan) present overviews of their projects to the Moore-Sloan community, CDS affiliated students and professors, and other data science researchers


What is the CDS Leadership Circle? 5 Minutes with Yiran Xu

Medium, NYU Center for Data Science


from

Yiran Xu, the president of the CDS student-run organization, the Leadership Circle, shares her plans for the upcoming year

 
Tools & Resources



Natural Language Generation at Google Research

YouTube, Google Cloud


from

In this episode of AI Adventures, Yufeng Guointerviews Google Research engineer Justin Zhao to talk about natural text generation, recurrent neural networks, and state of the art research!


23rd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Halifax 2017

KDD 2017, VideoLectures.NET


from

hours and hours of video at VideoLectures.NET


Word embeddings in 2017: Trends and future directions

Sebastian Ruder


from

“This post will focus on the deficiencies of word embeddings and how recent approaches have tried to resolve them. If not otherwise stated, this post discusses pre-trained word embeddings, i.e. word representations that have been learned on a large corpus using word2vec and its variants.”


Manipulating Maintainers

Michael Snoyman


from

“Real title should be: how to get members of any open source community to be interested in helping you. But the given title is catchier.”


The Census Bureau has begun rolling out our datasets via APIs.

U.S. Census Bureau


from

In alignment with the Digital Government Strategy, the http://videolectures.net/kdd2017_halifax/Census Bureau is offering the public wider access to key U.S. statistics. The Census application programming interface (API) lets developers create custom apps to reach new users and makes key demographic, socio-economic and housing statistics more accessible than ever before.”

 
Careers


Postdocs

Postdoc | Social Media Collective



Microsoft Research New England; Cambridge, MA

Postdoctoral Researcher in Computational Social Science



Microsoft Research; New York, NY
Full-time positions outside academia

Scientific Software Developer



Brookhaven National Laborratory; Upton, NY

Leave a Comment

Your email address will not be published.