SDS Lab students and alums enjoy spate of success stories
Three PhD students of the Sustainability and Data Sciences Laboratory (SDS Lab), viz., Kevin Clark, Babak Fard, and Thomas Vandal, successfully defended their dissertations last month.
Vandal received the College of Engineering best dissertation award among graduating PhD students. His research in deep (machine) learning and climate, which included collaborators from NASA Ames and the startup risQ, received a runner-up best paper award at ACM KDD 2017 in the applied data sciences track. His research on uncertainty quantification with Bayesian Deep Learning has been accepted in the ACM KDD 2018 research track. His research and mentorship inspired the research led by COE undergraduate student Matthew Mage at NASA Ames in deep learning and remote sensing.
Clark developed resilience approaches to multiscale aviation systems based on network science and system dynamics methods. His ongoing work at VOLPE and collaborations with SDS Lab PhD student Udit Bhatia helped in the process. Bhatia and Clark, along with SDS Lab PI Ganguly, are now collaborating with the Pacific Northwest National Laboratory in cyberthreats to aviation, especially in the context of interdependent network of networks formed by communications and transportation systems.
Fard’s work with AGU Thriving Earth Exchange and the town of Brookline on public health impacts of urban heat waves had received the first prize at the AGU Virtual Poster presentation last year. Fard has been collaborating with Bhatia and former SDS Lab postdoctoral associate Hanieh Hassanzadeh to develop this work further for urban climate risk management.
Bhatia has co-authored a textbook on Critical Infrastructures Resilience, and has also been helping run the SDS Lab by taking on many of the roles of a Lab PI, such as helping mentor graduate and undergraduate students, developing curricula, working on study abroad programs, and helping manage Lab finances. SDS Lab graduate students Mary Warner and Bharat Sharma passed their PhD qualifying exams in the last couple of months. Warner has dual affiliations with the SDS Lab and the Global Resilience Institute at Northeastern, and her dissertation is at the intersection of policy and engineering. Sharma and SDS Lab PhD student Shashank Konduri have been interning at the Oak Ridge National Laboratory, and are about to continue there under their Graduate Opportunities program. Konduri has been working in the broad area of food security, including under climate change, based on tools such as remote sensing, data analysis, network sciences, and models. Sharma has been working on carbon cycle extremes and climate drivers. The most recent PhD student of the SDS Lab, Kate Duffy (who started in 2017 Fall), is planning to intern at NASA Ames from June through August of 2018.
SDS Lab undergraduate student research interns Lindsey Bressler and Amina Ly graduated this May: the former will join the Peace Corps in Tanzania and the latter Stanford University as a graduate student. Masters student Summer Zacca, a Math major, just completed her research at the SDS Lab in extremes and nonlinear dynamics. SDS Lab PhD alum Evan Kodra and the PI Ganguly are two of three co-founders of the startup risQ which just received Phase II funding from the NSF SBIR program. Former SDS Lab postdoctoral associate Poulomi Ganguli received a faculty offer from the Indian Institute of Technology (IIT) at Kharagpur, the first and the largest IIT, while she along with former SDS Lab PhD student Devashish Kumar and PI Ganguly were cited as the authors of a “Top 100” paper in Nature’s Scientific Reports. Vandal’s KDD 2017 paper was highlighted in a comment in the journal Nature, while another paper based on research led by the University of Minnesota, where Ganguly was a co-author, was cited in Nature news, both as examples of pioneering work at the intersection of machine learning and climate. Bhatia and Konduri contributed significantly to the first international conference on the Networked Digital Earth at IIT Kharagpur this March, where Ganguly was a primary organizer and which saw speakers from five countries and a hundred and fifty odd attendees.
Ganguly is also a co-organizer of the fragile earth (FEED) workshop on theory-guided machine learning at ACM KDD 2018. He has also been invited to be a lead author of a cross cutting section on Artificial Intelligence (AI) for the US National Climate Assessment, and to join a review panel of the United Nations Environment Programme. Following the publication of their first textbook, Bhatia and Ganguly have started on their second textbook project, which will take an interdisciplinary and an applied perspective on time series and geospatial data analysis. Bhatia and Ganguly will also lead a study-abroad Dialogue of Civilizations program on climate change science and policy to Peru and Brazil this May and June. This program, which has been running in various parts of the world since 2014, will include an interdisciplinary group of thirty undergraduate students this year, including two student mentors, Yara Gosula and Francisco Franco Bulhoes Mendes. A textbook project on climate science and engineering by Bhatia, Ganguly, and collaborators, is also under discussion with the publisher. The three textbooks mentioned here covers, or will cover, the materials in the three courses Ganguly teaches on campus in the regular Spring and Fall semesters.