NYU Data Science newsletter – May 18, 2016

GROUP CURATION: N/A

New York City has been a Big Data pioneer for decades. In the early 1990s, the city launched the CompStat data-driven policing system, so that, in the words of former NYPD chief Lou Anemone, officers could stop “just running around answering 911 calls” and start analyzing patterns to prevent crime. Thanks in part to CompStat, major crimes in the city have since fallen by 80 percent.

During the Michael Bloomberg mayoral years, the city used data to pinpoint dangerous intersections and driving habits, cutting traffic deaths by nearly a third. Today, thanks to advances in data-storage capacity as well as the ubiquity of smartphones and broadband access, New York has an unprecedented number of facts to analyze and act upon, CompStat-style, across all areas of government — from building inspection to noise reduction.

But while it can shine a brighter light on problems and give citizens and government new tools with which to understand them, Big Data can’t solve the problems themselves. For that, we still need old-fashioned political will.

SAP believes the next technology adoption phase for businesses will be around how they can use intelligent applications to assist them with their operations.

Also, in enterprise software:

Salesforce CEO: I see an ‘AI-first world’ (May 18, Business Insider)

Everyone agrees that housing in San Francisco is expensive, and that the high costs are hurting the city. But there is a lot of disagreement about why the rent is so expensive, and what to do about it.

Sonja Trauss of SFBARF has argued that costs are high because there is not enough housing to go around and that the answer is to build more. Tim Redmond of 48 Hills has argued that building more housing would make the problem worse because the people who would move into it are likely to be wealthy newcomers whose demand for services will increase low-income employment, putting further pressure on older, lower-cost housing.

Who is right? Is anyone?

Indiana University data scientists will gather experts from across the globe in the nation’s capital starting today to review opportunities and challenges associated with the use of big data, visual analytics and computational models to advance public policy decisions related to science, technology and innovation.

The Modeling Science, Technology and Innovation Conference takes place May 17 and 18 at the National Academy of Sciences in Washington, D.C. The two-day event is funded by the National Science Foundation’s Science of Science and Innovation Policy program, the Cyberinfrastructure for Network Science Center, the IU Network Science Institute and industry.

The event is organized by Katy Börner and Staša Milojevi? of the IU School of Informatics and Computing’s Department of Information and Library Science and the IU Network Science Institute. The welcome and opening remarks will be presented by C.D. “Dan” Mote Jr., president of the National Academy of Engineering.

Predicting outcomes and, crucially, developing psychiatric drugs has proven exceedingly difficult in recent decades. Inadequate animal models have been a major stumbling block, researchers say. … “It’s a very exciting time,” says Guoping Feng, a neuroscientist at the Massachusetts Institute of Technology (MIT) in Cambridge, Massachusetts. “Between the technology development and the genetic findings, this is the first time that we’ve been able to begin digging deep into the causes and neurobiology of these disorders.”

Metadata. It’s an obscure data science term that was unknown to most people until 2013, when they learned that the U.S. National Security Agency (NSA) is harvesting vast amounts of it from telephone calls. Government officials have downplayed the sensitivity of such data, but a crowdsourced study of phone metadata now finds that highly revealing information can be gleaned from a simple list of who called whom.

NSA’s intrusion into citizen’s private lives may have roiled academics, but it has remained unclear what the spy agency was learning from phone metadata. A White House spokesperson reassured the public in 2013 that the metadata harvesting “does not allow the government to listen in on anyone’s telephone calls,” leaving privacy intact. Ever since then, a trio of computer scientists from Stanford University in Palo Alto, California—Jonathan Mayer, Patrick Mutchler, and John Mitchell—has been harvesting phone metadata themselves to see what can be revealed.

If you are trying to figure out what impact the new “Pascal” family of GPUs is going to have on the business at Nvidia, just take a gander at the recent financial results for the datacenter division of the company. If Nvidia had not spent the better part of a decade building its Tesla compute business, it would be a little smaller and quite a bit less profitable.

In the company’s first quarter of fiscal 2017, which ended on May 1, Nvidia posted sales of $1.31 billion, up 13 percent from the year ago period, and net income hit $196 million, up 46 percent over the same term. These are the kinds of growth numbers that all IT vendors like to show to Wall Street, especially with profit growth significantly outpacing revenue growth.

The datacenter portion of Nvidia, which it only started reporting on separately last year and for which it has given two years of financial results since it has become materially relevant, is growing much faster than the overall business.

Imagine you are driving on a highway late at night when a big-rig truck closes in behind you. You relax because it is keeping a safe distance and seems to be obeying the speed limit. Now imagine that truck is driving itself.

Despite Silicon Valley’s enthusiasm for self-driving cars, it could be years before there are many of them on the road. But autonomous 18-wheelers? One start-up is betting that is a different matter.

After trying without success more than a decade ago to set up preprint servers — where academics share their papers before peer review — science-publishing giant Elsevier is now buying one. It is paying an undisclosed sum for the Social Science Research Network (SSRN), one of the world’s most popular repositories of research in economics, law and the social sciences.

Also, in preprints:

It’s the Data, Stupid: What Elsevier’s purchase of SSRN also means (May 18, Savage Minds blog, Christopher Kelty)

Elsevier Acquires SSRN (May 17, The Scholarly Kitchen blog, Roger C. Schonfeld)

Preprints for the life sciences (May 20, Science; Jeremy Berg et al.)

The so-called cognitive revolution started small, but as computers became standard equipment in psychology labs across the country, it gained broader acceptance. By the late 1970s, cognitive psychology had overthrown behaviorism, and with the new regime came a whole new language for talking about mental life. Psychologists began describing thoughts as programs, ordinary people talked about storing facts away in their memory banks, and business gurus fretted about the limits of mental bandwidth and processing power in the modern workplace.

This story has repeated itself again and again. As the digital revolution wormed its way into every part of our lives, it also seeped into our language and our deep, basic theories about how things work. Technology always does this.

Facial recognition software is quickly becoming one of the most hotly contested fronts in the debate over online privacy.

Sites such as Facebook use facial recognition to enable users to tag their friends in photographs with greater speed and accuracy, but the technology has far-reaching ramifications for policing, security, and cyber-bullying. It is being challenged in federal court and is already feeding a cottage industry of new technology aimed at thwarting it.

Smarter and more adaptive machines are rapidly becoming as much a part of our lives as the internet, and more of our decisions are being handed over to intelligent algorithms that learn from ever-increasing volumes and varieties of data.

As these “robots” become a bigger part of our lives, we don’t have any framework for evaluating which decisions we should be comfortable delegating to algorithms and which ones humans should retain. That’s surprising, given the high stakes involved.

I propose a risk-oriented framework for deciding when and how to allocate decision problems between humans and machine-based decision makers.

Also, by Professor Dhar:

How can we control intelligent systems no one fully understands? (May 16, TechCrunch)

Instant Data Collaboration

Drag and drop your files to create a repository

It’s free to host public data, and free to store 100MB of private data

Michael Armbrust and Tathagata Das explain updates to Spark version 2.0, demonstrating how stream processing is now more accessible with Spark SQL and DataFrame APIs.

First to the naming; it basically is an arbitrary condensation of “R + Google Cloud Vision API”. I wonder why google chooses to mix google with vision. In my opinion it sounds pretty much like “to goggle with vision”, which makes limited sense. For the functionality; the package enables convenient Image Recognition, Object Detection, and OCR using the Google’s Cloud Vision API.