Ishaun got his earliest taste of research in high school, when he worked with a theoretical particle physicist to understand how nuclear matter binds. This motivated him to study physics, mathematics, and computer science at MIT. After graduating, he found his way back to particle physics, joining the CYGNUS experiment at Frascati National Laboratory (LNF) near Rome, Italy. The goal of CYGNUS is to identify the galactic origin of dark matter via a gas detector. Synthesizing tools from all three disciplines, Ishaun developed a method to analyze images from CYGNUS.
“My supervisor invited me to LNF to fill a void in the lab group’s existing knowledge of computer science, so I had enormous leeway to exercise creativity and establish my own research plan,” Ishaun said of his MISTI experience. “I decided how to approach the problem of analyzing CYGNUS data, and how to implement my approach. This flexibility was also a challenge, since I had to do all of my work independently.”
Researching modern clustering algorithms extensively, Ishaun applied the HDBSCAN algorithm to his data. This resolved some qualitative problems with the previous clustering method and resulted in a substantial quantitative improvement in runtime and clustering metrics, allowing him to identify particle tracks in the gas detector photographs. He also wrote software to analyze the geometric and topological features of the particle tracks to aid physicists’ data analysis.
Ishaun recognized MISTI as a cultural experience as well as one for research. “I took the train to nearby Rome to sightsee, visit museums, and taste Italian street food.”
Following this MISTI experience, Ishaun will travel to Munich, Germany to begin a DAAD research grant. As with his work in Frascati, Ishaun will explore the intersection of physics and computer science by studying bounds on the performance of quantum computational machine learning algorithms. He expects that the independence he was granted while pursuing his MISTI research will be useful in Germany, as well.