Philip B. Graff, Ph.D.

I am a Data Scientist with a Ph.D. in Physics from the University of Cambridge. I studied Physics and Mathematics at the University of Maryland, Baltimore Count (UMBC). I continued my studies at the University of Cambridge on a Gates Cambridge Scholarship and joined Queens' College. During my Ph.D., I worked on Bayesian inference for gravitational wave detection and parameter estimation. I applied these techniques to the training of neural networks through Bayesian optimization of the weights. Primarily, I worked on real and simulated data for the Laser Interferometer Gravitational wave Observatory (LIGO) project. Dissertation

Upon completing my Ph.D., I began a NASA Postdoctoral Program Fellowship at the Goddard Space Flight Center in Greenbelt, MD. I advanced and expanded the application of Bayesian inference and machine learning to the analysis of LIGO data. I brought the application of machine learning to analysis of other astronomical data, namely short Gamma-ray bursts as detected by the Swift satellite.

I joined the Johns Hopkins University Applied Physics Laboratory (APL) in 2015. At APL, I work on scalable analytics and data science. I focus largely on the development and application of graph-based analytics, utilizing the graph as a flexible and scalable data store.