Carnegie Mellon University: McWilliams Center Renamed to Reflect Expanded Research Scope
Tiziana Di Matteo(opens in new window), professor of physics(opens in new window), is leading efforts to unravel the origins of the universe and decode fundamental principles that govern its functioning. As director of Carnegie Mellon’s University’s McWilliams Center for Cosmology and Astrophysics(opens in new window), she’s making gravitational waves.
Astrophysics is now part of the center’s name, reflecting its new research direction.
“The name change reflects the broadening not only into more traditional astronomy but also the field of gravitational waves. Cosmology is the study of the universe as a whole. It answers the story of our origins and how everything starts,” Di Matteo said. “When we think of astrophysics, we think of understanding the components of the universe: How do galaxies form? What are black holes doing? How do the components of the universe behave and grow?”
Advances in telescopes and gravitational-wave detectors have created a data revolution in astronomy. Large-scale spectroscopic surveys such as the Dark Energy Spectroscopic Instrument(opens in new window) (DESI) survey, generate information on the scale of petabytes.
“Black holes are one of our greatest intellectual challenges and to reconcile our understanding of the largest structures in the universe with the smallest and other phenomena, we require data and information from many sources,” Di Matteo said.
Making sense of trillions of points of data requires a team effort. The McWilliams Center’s research lies at the crossroads of astrophysics, data science, artificial intelligence (AI) and machine learning. McWilliams Center members collaborate across disciplines not only within Carnegie Mellon’s Department of Physics but also in the Department of Statistics and Data Science(opens in new window) and Machine Learning Department(opens in new window).
Connecting Astrophysicists and AI
The connections between astrophysicists and AI researchers at Carnegie Mellon have been ongoing for decades, said Scott Dodelson(opens in new window), head of the Department of Physics and also the principal investigator for the National Science Foundation AI Planning Institute for Data-Driven Discovery in Physics(opens in new window).
“From analyzing data being collected from the entire night sky to modeling individual proteins in a cell, physics provides complex use cases and profound problems for researchers to consider as they advance AI techniques,” Dodelson said. “Carnegie Mellon is known for our expertise in computer science and artificial intelligence. The McWilliams Center is already making significant contributions to large collaborations because of our interdisciplinary approach.”
Physics Professor Rupert Croft(opens in new window) has helped organize conferences showcasing AI-assisted research across disciplines. He said that foundational work in both physics and computer science moves quickly, so researchers in diverse areas need to communicate and collaborate.
“For cosmology and astrophysics, the next step in AI simulations is to incorporate more of the physics,” Croft said. “Right now, we use gravity and dark matter. In stars and galaxies there is gas and dust, and those can be difficult parts to add to models. AI provides a framework for doing this quickly in the future.”
How big is a petabyte?
One petabyte equals 1,000 terabytes. In 2022, the Library of Congress managed 21 petabytes of digital collection content, comprising 914 million unique files. Other ways of looking at it, one petabyte could hold about 1,000 4K feature-length movies or is the equivalent of about 500 billion pages of printed text.
Physics Professor Rachel Mandelbaum(opens in new window) is a leading expert on developing analysis tools and techniques in cosmology. She is leading research to develop software to analyze large datasets generated by the Legacy Survey of Space and Time(opens in new window) (LSST), which will be carried out by the Vera C. Rubin Observatory.
Through the LSST, the Rubin Observatory, a joint initiative of the NSF and the Department of Energy, will collect and process more than 20 terabytes of data each night — up to 10 petabytes each year for 10 years — and will build detailed composite images of the southern sky.
“We’ll have all the data we need to make major discoveries about the origin of our solar system, the Milky Way, and the evolution of the universe, but we don’t have the software,” Mandelbaum said. “Many of the LSST’s science objectives share common traits and computational challenges. If we develop our algorithms and analysis frameworks with forethought, we can use them to enable many of the survey’s core science objectives.”
Multi-messenger Astronomy Unlocks Data Secrets
Di Matteo said the McWilliams Center is becoming a leader in Multi-messenger astronomy, a secret weapon for making sense of a sea of data.
“Through research efforts in big data and simulation modeling, Carnegie Mellon’s McWilliams Center is solidifying itself as a leader in this emerging field,” Di Matteo said. Recent faculty hires in astrophysics, such as Katie Breivik(opens in new window) and Antonella Palmese(opens in new window), are expanding research in his area.
“MMA has two main components of this research: big data, and simulation modeling,” Di Matteo said. “By integrating multiple streams of information, encompassing light, particles, and gravitational waves, the ultimate vision in MMA is to foster a deeper understanding of the universe’s most enigmatic phenomena.”
Breivik, who joined Carnegie Mellon in 2023 as an assistant professor, studies how binary stars evolve from birth to death. The phenomena — two stars orbiting a common center of mass — impact almost all aspects of astronomy, from shaping galaxies to making black holes. She teases information from datasets on stellar objects such as black holes, stellar cores and normal stars to develop simulations of binary star interactions. Along with machine learning experts, Breivik combines these models with data from gravitational-wave and electromagnetic surveys.
“We have a whole faculty that share a wide range of tools that range from theory to simulation to computation and are forging a path that actually develops new methods with each of these tools,” Breivik said. “When collaboration happens, you get magic way better than the sum of the parts.”
Simulations and computations are reinforced by observations by telescopes. McWilliams Center researchers have been involved in large telescope projects including the upcoming LSST and NASA’s James Webb Space Telescope(opens in new window) and upcoming Nancy Grace Roman Space Telescope(opens in new window).
Palmese, who joined Carnegie Mellon in 2022 as an assistant professor in physics, collaborates on international projects such as the Dark Energy Survey(opens in new window), DESI and the Laser Interferometer Space Antenna(opens in new window).
Palmese uses optical telescopes and sky surveys to find the light from catastrophic events giving rise to ripples of space-time that can be detected from collisions between neutron stars or black holes. To do this, she looks for evidence of transients, time periods where astronomical objects change over seconds, hours, or decades, within photometric and spectroscopic datasets.
“We have to deal with these very large datasets, whether it’s from galaxy surveys or sky surveys that are imaging the sky over and over again,” Palmese said. “You need big data techniques and machine learning to help you analyze those data and analyze it fast.”
Connections in Pittsburgh and Beyond
The McWilliams Center has access to resources at the Pittsburgh Supercomputing Center(opens in new window) (PSC), a joint collaboration between CMU and the University of Pittsburgh. This includes the dedicated cluster known as Vera.
“Having access to the Vera cluster has enabled a lot of the science that we’re able to do,” Palmese said.
The McWilliams Center also recently acquired direct telescope access on the South African Large Telescope(opens in new window) (SALT) with a mixture of internal funds. Researchers will use the time to rapidly observe explosions following the mergers of compact objects discovered through gravitational waves as well as provide students with crucial observing training.
McWilliams postdoctoral researcher Brendan O’Connor worked with Palmese on a McWilliams seed grant to purchase time on SALT, which will be in part to study cosmic explosions. One target he wants to study is the collision of two neutron stars, which produces a rare, short-lived transient, referred to as a kilonova. He said this is a developing area of astrophysics.
“Even a few observations of these rare objects can have a high impact at this point,” O’Connor said.
Postdoctoral fellows have been a part of the McWilliams Center since its beginning.
“Because of the McWilliams endowment we are competitive when it comes to being a destination for top candidates, who are, in turn, sought after for faculty positions,” Di Matteo said. “It’s a prestigious fellowship, and their success has been tremendous. They help make advancements possible. Postdocs ignite the place with new ideas and excitement.”
CMU Remembers Trustee Bruce McWilliams
The McWilliams Center for Cosmology and Astrophysics was established in 2007 with a gift from Bruce McWilliams, a Carnegie Mellon trustee and triple physics alumnus, who passed away in September 2023.
“Bruce was more than a benefactor,” Di Matteo said. “He played a pivotal role in shaping the academic community of the center, nurturing its growth with the same passion he held for the cosmos and physics.”
A resident of Silicon Valley, he was a passionate advocate and supporter of Carnegie Mellon.
McWilliams’ friends and colleagues are working to establish an endowed fellowship in his memory to support graduate students in the Mellon College of Science. For more information or to make a contribution, contact Nancy Felix(opens in new window), associate dean for advancement and strategic initiatives.