Pavlos Protopapas is the Scientific Program Director and Lecturer at the Institute for Applied Computational Science at the Harvard John A. Paulson School of Engineering and Applied Sciences.
As Scientific Program Director, he directs the educational program of the institute, overseeing the curriculum of the master’s programs in computational science and engineering and data science, teaching core data science courses, and mentoring master’s students and advising Ph.D. students in related fields.
Protopapas’ research is at the intersection of astronomy and data science. During his career, he has applied machine learning, deep learning, and statistics to gain knowledge about the cosmos. Currently, he is developing complete and diverse training sets to be used in the application of machine learning to astronomy, using deep transfer learning, generative adversarial networks (GANs), and deep neural networks. His research topics include many areas of machine learning and data science, especially classification, probabilistic graphical modeling, and time series analysis.
Protopapas has a Ph.D. in theoretical physics from the University of Pennsylvania, and a BS in physics from Imperial College, London. From 2005 to 2013, he was at the Harvard-Smithsonian Center for Astrophysics. Previously, he was the associate director of the National Scalable Cluster Project, initiating one of the first ever attempts of using large-scale distributed computing on a grid-like model.