Rohan Arni, 17, used deep learning to study mysterious space signals with 98% accuracy; now he is a US Regeneron STS finalist

Rohan Arni, a 17-year-old student at High Technology High School in Lincroft, New Jersey, is among the 40 finalists of the 2026 Regeneron Science Talent Search. His project uses deep learning to classify fast radio bursts (FRBs), mysterious millisecond-long flashes of radio waves from space. Using data from the Canadian Hydrogen Intensity Mapping Experiment, his model achieved 98% accuracy in distinguishing repeating and non-repeating FRBs, offering astronomers a potential new tool to study their origins.