Matteo Paz, the 18-year-old son of Amy and Pedro Paz, is making waves far beyond the halls of Pasadena High School where he studies. As president and founder of his school's research club, he mentors fellow students in science competitions, but his own scientific achievements have recently captured global attention.
A Stellar Discovery from a Computer Screen
Unlike traditional astronomers who peer through telescopes, Paz made his groundbreaking discovery while sitting at a computer, meticulously analyzing an ocean of numbers collected high above Earth. The young Pasadena student has gained recognition for identifying an astonishing 1.5 million previously unknown variable objects in space, all drawn from more than a decade of observations by NASA's WISE space telescope.
What began as a science competition project quickly expanded beyond its original scope. By employing artificial intelligence and advanced mathematical techniques, Paz developed a method to uncover patterns hidden within one of the largest astronomical datasets ever assembled. His findings reveal black holes, newborn stars, and distant cosmic explosions, quietly expanding our scientific understanding of a dynamic and ever-changing sky.
Mining Infrared Data for Cosmic Treasures
NASA's WISE and subsequent NEOWISE missions have been scanning the entire sky in infrared light for over ten years, accumulating nearly 200 billion individual observations. This created such an enormous dataset that most of it remained unexplored until Paz recognized an opportunity.
Rather than examining objects individually, the innovative teenager designed a system capable of scanning the complete archive for changes in brightness over time. These fluctuations often indicate rare or highly energetic cosmic events. His approach transformed raw data into something manageable and searchable on a scale few individuals have ever attempted.
Life Beyond the Stars: Leadership and Mentorship
Paz's accomplishments extend far beyond astronomical research. He actively participates in school leadership, having served on his district's first unified student council and as a student assembly representative to the school board. His commitment to education led him to establish a research club specifically to support other students entering science competitions.
Additionally, Paz runs a program called Money Matters, which introduces middle school students to fundamental financial literacy concepts. These diverse efforts complement rather than compete with his research, suggesting a pattern of sustained curiosity rather than a single moment of achievement.
The VARnet Machine Learning Model
At the heart of Paz's project lies VARnet, a sophisticated machine learning model he developed himself. This innovative system combines signal processing, wavelet analysis, and deep learning techniques to recognize faint or irregular patterns in astronomical light curves.
The model was trained using simulated data and known infrared variables before being tested on actual observations. According to published research, VARnet processes each source in a fraction of a millisecond when utilizing modern GPUs. This remarkable speed enabled the analysis of the entire NEOWISE single exposure database, ultimately identifying 1.9 million variable objects, most previously uncatalogued.
What the New Catalogue Reveals
The newly identified objects encompass a broad spectrum of cosmic phenomena. Some represent supermassive black holes actively consuming matter at galactic centers, while others include young stars in formation or supernovae briefly flaring before fading from view.
Variable objects hold particular scientific value because they reveal motion, growth, or sudden change in the universe. Paz's comprehensive catalogue provides researchers with a more complete map of this activity across the infrared sky, offering a valuable starting point that eliminates the need to search blindly through raw observational data.
Educational Pathways to Discovery
Paz's academic preparation at Pasadena High School included advanced mathematics through the district's Math Academy, where he completed high-level coursework years ahead of schedule. His interest in artificial intelligence developed through an elective that blended coding, theoretical concepts, and formal mathematics.
According to Caltech experts, this educational background helped Paz recognize that large, well-organized datasets present ideal opportunities for machine learning applications. The project was conducted under NASA funding, with Paz working as a staff researcher rather than a casual intern, reflecting the professional significance of his contributions.
