Jonathan Richardson, an assistant professor of physics and astronomy; and Vagelis Papalexakis, an associate professor of computer science and engineering; have received a two-year, $300,000 grant from the National Science Foundation, or NSF, to develop novel machine-learning methods capable of analyzing the physical origins of noise in LIGO, the Laser Interferometer Gravitational-wave Observatory based at sites in Louisiana and Washington.
LIGO detects gravitational waves — distortions in the fabric of space — using detectors that employ high-power laser beams, offering a new way to observe the universe.
“Success in this project will improve the operational stability of the detectors and increase their astrophysical range, which could advance scientific discovery,” said Richardson, the grant’s principal investigator. Papalexakis, the grant’s co-principal investigator, will help bridge the gap between physics and machine learning.
Richardson explained that despite a series of upgrades and improvements, the LIGO detectors suffer from noise whose origin is largely unknown and whose presence limits the detectors’ astrophysical reach.
His research team, comprised also of UCR graduate and undergraduate students, will develop open-source tools for understanding and detecting the noise. Such tools can significantly improve the sensitivity, data quality, and operational stability of LIGO and future facilities, and could possibly increase detection rates of mergers of the most massive stellar black holes — extraordinarily powerful and energetic events that produce the lowest-frequency gravitational waves — by more than a factor of six.
Prior to joining UCR, Richardson served as a LIGO Laboratory postdoctoral scholar at the California Institute of Technology from 2017-21. Recently, he received a grant, also from NSF, to develop new laser wavefront control capabilities for gravitational wave detectors. The new grant is awarded through a special NSF program, Advancing Discovery with AI-Powered Tools, focused on applying artificial intelligence/machine learning to address problems in physics.