A team of students from the UMass Amherst College of Information and Computer Sciences (CICS), led by doctoral student Ryan McKenna and including graduate students Joie Wu, Arisa Tajima, Brett Mullins, Siddhant Pradhan and Cecilia Ferrando, recently won $43,000 in prize money in the 2020 National Institute of Standards and Technology (NIST) Differential Privacy Temporal Map Challenge.
The Differential Privacy Synthetic Data Challenge was established by NIST in 2018 to help advance research for public safety communications technologies while assuring data privacy, including helping first responders gain insights from public safety data. This year's challenge was to develop algorithms to anonymize large sets of temporal map data, which are used by the public safety community during natural disaster responses, epidemic tracking, and civic planning.
The challenge featured a series of three coding sprints, which required competitors to apply differential privacy methods to temporal map data, including a year's worth of data from 911 calls in Baltimore and a set of millions of taxi trips in Chicago. The CICS team, calling themselves the Minutemen, took first place in Sprint 1, third place in Sprint 2, and second place in Sprint 3, resulting in a cumulative prize of $43,000.
“We are all differential privacy researchers, and found the temporal and geographical aspect of the challenge very interesting,” says McKenna. “In addition to the fun of the challenge, we are proud to contribute to the research that is helping to make these critically important datasets available to first responders.”
Read more about the challenge and the team's winning solutions.