High School Student Unveils 1.5 Million Potential New Astronomical Objects by Creating An A.I. Algorithm

Photo credit: www.smithsonianmag.com

The 18-year-old won $250,000 for training a machine learning model to analyze understudied data from NASA’s retired NEOWISE telescope

Margherita Bassi – Daily Correspondent

April 15, 2025 5:19 p.m.

A significant advancement in astronomy has emerged, led by an innovative researcher who has successfully harnessed artificial intelligence to uncover over one million celestial objects through an analysis of data from NASA’s NEOWISE telescope.

This remarkable achievement was documented in a study published in November in The Astronomical Journal. Notably, the sole author of this groundbreaking study is just 18 years old.

Matteo Paz, hailing from Pasadena, California, recently secured the grand prize of $250,000 in the 2025 Regeneron Science Talent Search. This esteemed competition, often perceived as the most prestigious in the nation for high school seniors in science and math, recognized Paz’s novel integration of machine learning within the field of astronomy. His advanced algorithm analyzed an astounding 200 billion data entries from NASA’s retired Near-Earth Object Wide-field Infrared Survey Explorer (NEOWISE), leading to the identification of 1.5 million new potential celestial entities.

“I was just happy to have had the privilege. Not only placing in the top ten but winning first place came as a visceral surprise,” Paz shared with Forbes’ Kevin Anderton. “It still hasn’t fully sunk in.”

Paz’s journey into astronomy was catalyzed by his participation in the Planet Finder Academy at the California Institute of Technology (Caltech) during the summer of 2022. There, he delved into astronomy and computer science under the mentorship of Davy Kirkpatrick, an esteemed astronomer and senior scientist at Caltech’s Infrared Processing and Analysis Center (IPAC).

Kirkpatrick had been utilizing data from the NEOWISE infrared telescope, which was launched in 2009 primarily to identify near-Earth asteroids and comets. However, the telescope also gathered valuable data on variable objects—celestial phenomena that emit varying light, such as exploding stars. Kirkpatrick envisioned searching for these rare entities within NEOWISE’s often-overlooked data.

“At that point, we were approaching 200 billion rows in the database of every NEOWISE detection recorded over more than a decade,” Kirkpatrick stated in a Caltech press release. “My goal for the summer was to explore a small segment of the sky to identify some variable stars and highlight these findings to the astronomical community, emphasizing the potential contained within the dataset.”

Instead of manually sifting through the data, Paz developed an A.I. model designed to detect subtle fluctuations in infrared emissions, which could signify the presence of variable objects. Following the summer program, he and Kirkpatrick collaborated to refine the model, which ultimately detected 1.5 million new potential objects, including supernovae and black holes.

“Before Matteo’s initiative, no one had attempted to analyze the entire 200-billion-row dataset to classify significant variabilities,” Kirkpatrick mentioned in correspondence with Business Insider. He revealed that Caltech researchers are already utilizing Paz’s newly established catalog of variable objects, named VarWISE, to investigate binary star systems.

“The variable candidates he has identified will be extensively researched,” commented Amy Mainzer, NEOWISE’s principal investigator at NASA, to Business Insider.

In terms of broader applications, Paz noted the potential versatility of his A.I. model, indicating that it could also be beneficial in other fields where data is time-dependent, such as financial markets and environmental monitoring. His keen interest in climate issues is further emphasized by his personal experience with wildfires in Los Angeles earlier this year, when his family was forced to evacuate due to the Eaton Fire, as reported by Forbes.

Alongside Paz, other innovative young scientists honored by the competition tackled diverse topics, including strategies for mosquito control, drug-resistant fungi, genetic research, and advancements in mathematics.

“The remarkable creativity and dedication of these students instills renewed hope for our future,” stated Maya Ajmera, president of the Society for Science, which administers the renowned award, in a statement. “Their ingenuity is leading to revolutionary solutions with the potential to transform our world and drive societal progress.”

Source
www.smithsonianmag.com