A California high school student has revolutionized astronomy by developing an artificial intelligence model that discovered over 1.5 million previously undetected space objects. Matteo Paz, a teenager who spent his summer researching at Caltech’s Planet Finder Academy, applied machine learning techniques to the massive dataset collected by the NASA telescope NEOWISE.
The project, led by Professor Andrew Howard and senior scientist Davy Kirkpatrick, aimed to analyze the data from NEOWISE, which was initially designed to track near-Earth asteroids. However, the mission also captured thermal emissions across the entire sky for over a decade, yielding vast amounts of data on distant cosmic objects.
Paz’s AI model, built in just six weeks, successfully scanned the dataset and flagged objects with variable brightness. The student’s advanced math education and background in computer science enabled him to work at a level typically reserved for college researchers.
Collaborating with other Caltech researchers, Paz refined his pipeline and overcame challenges associated with detecting subtle variations in infrared light. His findings were published in The Astronomical Journal, and a complete catalog of detected variable objects is scheduled for release in 2025.
The breakthrough highlights the potential applications of machine learning beyond astronomy, including time-domain studies in other fields such as finance or environmental monitoring. Paz, who is now working at IPAC while finishing high school, plans to continue refining his model’s capabilities and mentoring other students through the Planet Finder Academy.
Source: https://dailygalaxy.com/2025/04/i-mapped-the-invisible-an-american-high-school-student-turns-astronomy-on-its-head-with-his-ai-that-reveals-1-5-million-forgotten-space-object