Alexander Gray, assistant professor in Computational Science and Engineering, has received a CAREER Award from the National Science Foundation for his project, "Scalable Machine Learning for Astrostatistics."

The five-year award, which at nearly $600,000 is 50 percent larger than a typical CAREER Award, will continue Gray's work in astrophysics, in which he's been involved since the early 1990s. He has applied computational algorithms to study formation of galaxies, determine evidence for "dark energy" in the universe, and study the formation and morphology of galaxies.

The project that won him the CAREER Award will involve creating new data structures and algorithms to scale up existing machine learning techniques to the point that they can deal with the massive datasets involved in cosmological phenomena. Gray said he will specifically explore a new data structure called a "cosine tree" to help manage the computational bottleneck that results from massive datasets. As an educational component, he said the project will test a novel approach to Ph.D. curricula.

"Building upon the successful Threads model pioneered at Georgia Tech, the innovations center around extensions of its underlying ideas of contextualization and symphonic thinking," Gray said. "[The innovations will involve] Cross-Threads, a way to curricularize cross-disciplinary inspiration through a graduate analog of the Threads concept, representing a new kind of design for a Ph.D. program in computing."