- An AI model builder developed at the University of California, Irvine can autonomously design particle physics theories, a task traditionally performed only by humans.
- This approach enables efficient exploration of new theoretical frameworks and helps discover promising neutrino models for scientific research.
- In the future, the application may be extended to other theoretical model building problems.
- The National Science Foundation and the Department of Energy provided research support.
Irvine, California, July 9, 2026 —Physicists at the University of California, Irvine have developed an artificial intelligence system that can autonomously design theoretical physical models that were traditionally done by human theorists. This approach allows researchers to explore vast uncharted territory in particle physics theory and helps identify promising new explanations for neutrino behavior.
The system, called Autonomous Model Builder (AMBer), was developed by a research team led by Victoria Knapp-Pérez and Jake Rudolph, doctoral candidates in the Department of Physics and Astronomy at the University of California, Irvine. This research natural communication physics.
AMBeer uses reinforcement learning. Reinforcement learning is a type of artificial intelligence that learns through trial and error rather than following predefined instructions. As it explores potential particle physics theories, the system evaluates its own choices and improves them over time.
“Reinforcement learning is different from other types of machine learning, where the model predicts labels or finds patterns in data,” Rudolph said. “AMBeer’s RL framework allows you to learn about the space of theoretical models during exploration and effectively create your own training data as you explore promising models.”
The system builds a particle physics model by selecting a mathematical symmetry group, deciding which particles to include, and assigning how those particles behave under the selected symmetries. Each proposed model is evaluated based on how well it matches experimental data while minimizing the number of tunable parameters, which is an important measure. The predictive power of theory.
The researchers tested AMBER on a well-studied class of neutrino theory and demonstrated that it could reproduce known results. They then applied this system to a previously unexplored mathematical framework and identified new candidate models that may be worthy of further investigation.
Neutrinos are very small but non-mass subatomic particles, properties that cannot be explained by the standard model of particle physics. Developing a theory to explain the mass of neutrinos remains one of the field’s major challenges.
The researchers emphasized that the system is not designed to replace human physicists, but to assist them by narrowing down the vast scope of research. The theory is to narrow down the most promising candidates.
“AMBeer acts as a filter, providing human physicists with a more informed starting point to study the more complex behavior of neutrino models,” Knapp-Perez said.
Additional collaborators include Max Fieg, a former doctoral student at the University of California, Irvine, and now a postdoctoral fellow at Fermilab. Aishik Ghosh, former postdoctoral fellow at the University of California, Irvine, and currently a professor at the Georgia Institute of Technology. Daniel Whiteson, a professor of physics at the University of California, Irvine, oversaw the research. Jason Barrett, a doctoral student at the University of California, Irvine, in Whiteson’s group also contributed to the study.
This study used computational resources from the National Energy Research Center for Scientific Computing. Funding was provided in part by the National Science Foundation UC-MEXUS-CONACyT and the Department of Energy’s Office of High Energy Physics.
About University of California, Irvine: Founded in 1965, UC Irvine is a member of the prestigious Association of American Universities and ranks among the nation’s top 10 public universities worldwide. US News & World Report. The campus has produced five Nobel Prize winners and is known for its academic achievements, top-notch research, innovation and anteater mascot. Led by President Howard Gilman, the University of California, Irvine has more than 36,000 students and offers 224 degree programs. Located in one of the safest and most economically vibrant regions in the world, we are Orange County’s second largest employer, contributing $7 billion annually to the local economy and $8 billion to the state overall. For more information about the University of California, Irvine, visit www.uci.edu.
Media access: Radio programs/radio stations may use, for a fee, an on-campus studio equipped with a Comrex IP audio codec to interview UC Irvine faculty and experts, subject to availability and university approval. For more University of California, Irvine news, visit news.uci.edu. Additional resources for journalists are available at https://news.uci.edu/media-resources.
