08-04-2025
Teen Wins $250,000 For Discovering 1.5 Million New Space Objects
An 18-year-old student from Pasadena, California, placed first in the nation's oldest and most prestigious science and mathematics competition for high school students, winning $250,000.
Matteo Paz was one of the finalists who submitted an original research science project in this year's Regeneron Science Talent Search. The students represented 39 schools across 16 states and the finalists competed for nearly $2 million in prizes and awards. Paz's project analyzed a massive amount of astronomical data collected by NASA's WISE explorer telecope. Using an algorithm he developed, he sorted through the information and identified over a million new celestial objects in the night sky.
'Congratulations to the winners of this year's Regeneron Science Talent Search,' said Maya Ajmera, President and CEO, Society for Science and Executive Publisher, Science News. 'The remarkable creativity and dedication of these students bring renewed hope for our future. Driven by their ingenuity, these young scientists are developing groundbreaking solutions that have the potential to transform our world and propel society forward.'
Winning first place was a shock to Paz, who had no expectations after being named a finalist.
'I was just happy to have had the privilege. Not only placing in the top 10, but winning first place, came as a visceral surprise. It still hasn't fully sunk in,' remarked Paz.
The path to victory was not without its trials, as earlier this year, he and his family had been evacuated from their home in Pasadena due to the Eaton Fire. During the project's development, he faced challenges in the code or quirks within the data itself. Fortunately, Paz found the process of refining algorithms, and mathematical complexities deeply engaging. As he looked through the primary results, he began seeing unique and unstudied objects.
"What's exciting is that some variable phenomena I'm detecting don't have obvious explanations. These quirky, mysterious objects are exactly those that spark new science and physics," remarked Paz.
Launched in 2009, NASA's Wide-Field Infrared Survey Explorer (WISE) collected 200 billion lines of infrared astronomical data (200 terabytes). However, this data had not been cataloged. In his Regeneron Science Talent Search project, Matteo Paz developed waveform-based machine learning methods to detect and characterize variables within the data. These methods included a machine-learning algorithm dubbed VARnet.
Paz developed VARnet to detect variable objects specifically by combining machine learning with particular mathematical concepts. The first mathematical operation served to de-noise data while conserving short-timescale variation. The second operation, which Paz calls the Finite-Embedding Fourier Transform, serves to extract patterns in the data. Working together, these operations resulted in the identification of 1.9 million infrared variable objects, including 1.5 million new discoveries. Among those discoveries are objects like supermassive black holes, newborn stars, and supernovae.
Matteo Paz first place research poster at the 2025 Regeneron Science Talent Search. The first ... More complete infrared variability survey - detection and classification of 1.9 million objects
Listen to Paz explain his project in his own words
Learn more about Matteo Paz here.
After wrapping up the VarWISE project, Paz would like to continue his work in astronomy. He believes that, by using the discoveries from his project, we could properly measure the rate of expansion of the universe and challenge our understanding of its origin. However, he feels that finding funding over the next few years may prove difficult.
To learn more about all of this year's top 40 finalists, visit the Society for Science.
