image: Illuminating the Dark. Mysterious dark matter and neutrinos, formed moments after the Big Bang, are the focus of a new cosmological collaboration between USC, UC Riverside and the Carnegie Science Observatories.
Credit: Martzi C. Campos/USC School of Cinematic Arts
Only about 5% of the universe consists of particles and forces well understood by modern physics. The rest remains a puzzle — one that involves dark matter, an invisible substance that shapes galaxies and cosmic structures, as well as ubiquitous but poorly understood particles called neutrinos, which were produced moments after the Big Bang.
With a $4 million grant from the John Templeton Foundation, a team of scholars from USC, the University of California, Riverside, and the Carnegie Science Observatories will establish a research hub to better understand these mysterious components of the cosmos.
The new Lyman-Alpha forest Research Collaboration (LARC) will develop cutting-edge computer models to simulate the birth of structure in the universe, testing different theories about how galaxies form. Comparing these computer-grown models of the universe to observations of the real thing will help reveal quantum properties of the universe’s hidden constituents, including dark matter and neutrino particles.
Vera Gluscevic, associate professor of physics and astronomy at the USC Dornsife College of Letters, Arts and Sciences, will serve as the collaboration lead with Simeon Bird, associate professor of astrophysics at UC Riverside, as co-lead.
“At its core, LARC is designed to pursue the deepest questions about our universe and the essence of human knowledge: What is the fundamental nature of matter? What forces shape physical reality? What is the meaning of ‘discovery’ when AI and simulation guide the pursuit of truth?” says Gluscevic.
Bird says the collaboration assembles a uniquely qualified group of scholars to tackle these questions. “The most exciting part is the team, experts in a vast range of disciplines, including theoretical and observational astrophysics, philosophy, computer science, and even interactive data visualization” he says. “I am especially excited to work with computer scientists to make better simulations of the Universe, which will help us understand where it comes from and what it’s made of.”
To boldly go …
Among the main goals of the LARC team is to use pioneering observations of hydrogen gas in space, created by scientists Drew Newman and Gwen Rudie at Carnegie Observatories, to trace dark matter. “Thanks to new observations with large telescopes, we can now map the three-dimensional structure of intergalactic gas. With the LARC team, we’ll create new ways to use these exciting maps to learn about the hidden universe,” said Newman.
LARC also includes Aiichiro Nakano, professor of computer science, physics and astronomy, and quantitative and computational biology at USC Dornsife and the USC Viterbi School of Engineering, and Christian Shelton, professor of computer science and engineering at UC Riverside. Both use artificial intelligence techniques to recognize useful patterns in computer simulations, and speed them up. This enables astrophysicists to compare simulations to real data collected by telescopes and space missions.
In doing so, they raise an important philosophical question, one which the group also plans to tackle.
Historically, scientific understanding meant formulating theories simple enough to be grasped and explained by humans. Now, the sheer volume and complexity of astronomical data challenge this traditional approach, as many insights come from computer-generated models beyond direct human intuition. What does it mean to “understand” the universe when our knowledge comes from computer-generated models?
To dig into this conundrum, a team of philosophers, led by Dmitri Gallow, associate professor of philosophy at USC Dornsife, will examine how new scientific methods such as these may change our understanding of how we produce scientific conclusions and generate knowledge.
“Using AI to understand a system as complex as the universe poses significant rewards — but also significant risks,” Gallow says. “We need to think carefully about how to responsibly learn from AI.”
The art of science
The public will get a chance to try their hand at this cosmic exploration, as well. Martzi Campos, assistant professor of cinematic arts, and game lab research associate Sean Bouchard, both with the USC School of Cinematic Arts, will produce a 3D interactive visual, which will be displayed in the Visualization Lab on the Carnegie Observatories campus, as well as educational games for the public using the hub’s work as inspiration.
The wide scope of this project is in keeping with the ambitions of John Templeton himself, a maverick investor who launched the foundation with the aim of funding advances in scientific discovery and inspiring awe and wonder for the universe.
“Understanding the fundamental nature of dark matter and neutrino particles would present a seismic advance in science, likely opening whole new directions of research in particle physics and cosmology,” Gluscevic says. “And, if we can help redefine the formal process of ‘discovery’ in the era of AI and computer simulation, we may enable future research to proceed on firmer ground and at greater speed than ever.”