In a remarkable turn of events that challenges the way we think about space explorationa high school student in California has made a discovery that could reshape our understanding of the cosmos. Matteo Paz, a teenager with a sharp aptitude for computer sciencehas developed an artificial intelligence model that uncovered 1.5 million space objects previously unknown to astronomers. The objects were hidden in plain sight within a vast dataset collected by NASA’s NEOWISE missiona project designed to track near-Earth asteroids.
The Genesis of an Unexpected Discovery
Matteo Paz’s journey into deep space began during the summer of 2022 when he participated in Caltech’s Planet Finder Academy. The program, led by Professor Andrew Howardaims to provide high school students with direct exposure to advanced astronomical research. Paz was mentored by Davy Kirkpatricka senior scientist at Caltech’s Infrared Processing and Analysis Center (IPAC), and together, they began exploring the treasure trove of data from the NEOWISE mission.
Initially designed to observe and track asteroids near Earth, NEOWISE had spent more than a decade gathering infrared data that covered the entire sky. While the mission achieved its primary goal, it also gathered valuable information on distant objects—particularly those that exhibited unusual brightness variations over time. These variations, sometimes caused by quasarsbinary stars, or supernovae, were often difficult to pinpoint using traditional methods of analysis.
The Power of AI in Space Exploration
Faced with a dataset that was rapidly approaching 200 billion data points, the Caltech team had planned to analyze only a small portion manually. However, Paz, with his background in computer sciencemathematics, and programming, saw a different path forward. He proposed using machine learning to handle the massive dataset, a solution that no one had considered before.
In just six weeks, Paz built a Fourier and wavelet-based machine learning model designed to identify the faintest signals of brightness fluctuations in the infrared data. His efforts quickly paid off. The model began detecting subtle variations in light that had previously gone unnoticed. As the AI refined its ability to spot unusual patterns, it flagged potential new objects and activities in the vast expanse of space.
Overcoming Challenges and Mentoring the Next Generation
Paz’s breakthrough, was not just a technological achievement, but also a testament to the power of mentorship. Throughout the project, published in The Astronomical JournalPaz worked closely with a team of researchers at Caltech, including Shoubaneh Hemmati, Daniel Masters, Ashish Mahabal, and Matthew Graham. These experts provided critical support in refining the machine learning techniques that would eventually lead to the discovery of the hidden objects.
However, the journey was not without obstacles. The NEOWISE mission’s observation rhythms had limitations that made it difficult to detect certain kinds of variable objects—those that changed too slowly or too rapidly. Paz’s AI model, which could track these variations in brightness over time, proved instrumental in overcoming this challenge. His work has led to the detection of new stars and other phenomena, opening doors for further study and exploration.
Now, as a paid employee at Caltech, Paz continues to refine his model, pushing the boundaries of what his AI can achieve. He also works to mentor younger students in the Planet Finder Academyensuring that the next generation of researchers will have the tools and inspiration to explore space in new ways.
A Discovery That Could Change How We See the Universe
The catalog of 1.5 million new objects uncovered by Paz’s AI is set to be published in 2025, offering fresh insights into the evolution of distant stars and galaxies. These findings will allow astronomers to study the long-term behavior of celestial bodies that were previously invisible. While the discovery is rooted in space research, Paz sees the potential for broader applications of his model, including in fields like economics and environmental monitoring, where temporal data plays a critical role.
As a high school student with an exceptional understanding of machine learning and astronomical data, Paz’s success exemplifies the potential of young talent when combined with the right resources. His work also underscores the power of advanced technology, especially artificial intelligence, to uncover the unseen mysteries of our universe.
