Posted by Rob Knies

New York City

It’s been two years since Microsoft Research New York City was founded—and nine months since the lab established a permanent, bespoke home in the Flatiron District, smack dab in the middle of Silicon Alley. So how have things gone so far?

Pretty darn well, if you ask Jennifer Chayes, managing director of the New York City facility.

“In many ways,” she says, “the lab has become the top research group in data science worldwide.”

That’s one of the messages Chayes and colleagues will be delivering on May 5, when Microsoft Research New York City hosts an open house to show off its new space to an assemblage of select academics.

The event is the latest in a series of momentum-building developments since the lab opened in May 2012. Its researchers collaborate with Microsoft Research colleagues and with academia to advance the state of the art in various disciplines. Those efforts, Chayes says, have enabled the lab to build a full head of steam.

“We have combined the more conventional computer-science fields of machine learning and information retrieval with economics, sociology, and behavioral psychology—to stunning effect,” she says. “Working hand in hand, the computer scientists and the social scientists have made amazing strides in prediction engines, computational economics, crisis informatics, and online virtual laboratories for behavioral and social sciences.

“And they’ve only just begun!”

The academics in attendance for the open house will get a chance to learn about some of these advances, via three presentations from lab scientists and a panel discussion. The former include:

The panel discussion, titled Interactions Between the Academy and Silicon Alley in the Age of Data Science, will feature Chayes, along with a selection of the city’s other vibrant thought leaders:

  • Dan Huttenlocher, dean and vice provost of Cornell NYC Tech.
  • Kathleen R. McKeown, director of the Institute of Data Science and Engineering at Columbia University.
  • Clay Shirky, whose focus is on studying the societal effects of the Internet. He has a joint appointment at New York University (NYU), serving as associate arts professor in the Interactive Telecommunications Program and as associate professor in the Department of Journalism.

With the presence of world-renowned universities and a vibrant startup ecosystem, New York has become a data-science hub, and Microsoft Research New York City is an active participant. It helps that the lab is located close to NYU—and that Microsoft engineers working on Yammer and Skype are mere steps away.

“We’ve been incredibly successful in connecting with the local external research community,” Chayes says. “We have very strong ties to the three Bloomberg Applied Science initiatives—Cornell Tech, the Center for Urban Science and Progress at NYU, and the Institute for Data Sciences and Engineering at Columbia.

“We’ve also been interacting deeply with other groups, including the Data Science Institute, the Media Culture and Communications Department, and the Interactive Telecommunications Program, all at NYU. In fact, in late April, we had our first Microsoft Research New York City Data Science Seminar Series, in which we brought together people from all these institutions for an evening networking event on data science.”

There are other ways to measure the lab’s burgeoning influence:

  • Headcount: Microsoft Research New York City was launched with a dozen newly hired researchers. Now, that number has increased 58 percent, to 19, which doesn’t include Chayes or the three postdocs currently based at the lab.
  • New hires: Of course, the talent component is not just about numbers. It’s mainly about the quality of the hires. Take Robert Schapire, for example. Formerly a professor at Princeton University, he brings to the lab a reputation as one of the world’s top machine-learning experts. Or consider Hanna Wallach, who came to Microsoft from the University of Massachusetts Amherst, where she was a key player in that school’s Computational Social Science Initiative.

And then there are the researchers who joined the company as members of Microsoft Research New England, also managed by Chayes, but have since relocated to New York. Their names might sound familiar: Crawford and danah boyd.

  • Prediction modeling: Rothschild has received lots of attention for his work, particularly in the wake of his astonishing accuracy in picking the winners in the 2012 U.S. presidential election and the 2014 Academy Awards. What’s he up to now? Predicting the likelihoods for victory of the 37 nations competing in the 2014 Eurovision Song Contest.
  • Stellar papers: Sharad Goel, Jake Hofman, and Watts took social-media research to a new level with their paper The structural virality of online diffusion, written in conjunction with Ashton Anderson of Stanford University. Another game-changing research paper, The Cost of Annoying Ads—written by Dan Goldstein and Siddharth Suri of Microsoft Research New York City, along with Google’s Preston McAfee—described for the first time how to quantify the disincentives web publishers receive when they run an irksome display ad.
  • Community initiatives: Chayes already mentioned the Center for Urban Science and Progress, but that’s not the only work Microsoft Research is pursuing that delivers direct benefits to its New York neighbors. For instance, Kati London’s HereHere NYC project enables neighborhoods to generate aggregated opinions based on public data.

Altogether, the achievements from Microsoft Research New York City over its initial two years have been many and varied—and given its location and the milieu in which it operates, that momentum seems certain to continue to grow.

“We’re thrilled,” Chayes says, “that New York City has such a vibrant community in data science, both in academia and in Silicon Alley.”