Collaborative Research: CDS&E: Data-Driven Discovery of Neural ODE Dynamics, Astrophysical Models, and Orbits (Neural ODE DynAMO)

Scientists recently detected gravitational waves from black holes and neutron stars orbiting each other. These waves, which were first predicted by Albert Einstein, are like ripples in space-time created by the movement of massive objects in distant galaxies. Current models of these gravitational waves focus on two-body systems, like a pair of black holes or neutron stars orbiting each other, without considering the effects of their astrophysical environments. The main scientific challenge is incorporating these complex environmental interactions into the models. This award will fund an interdisciplinary team from multiple institutions to use new machine learning advancements to tackle these challenges. The resulting models and machine learning techniques will allow researchers to study powerful collisions of binary black holes in extreme environments. Moreover, this research project will engage the public through outreach activities and train diverse students with strong backgrounds in science, technology, engineering, and math, preparing them for careers requiring technical and computational skills. This research team's previous work introduced gravitational waveform inversion (GWI), a machine-learning technique for discovering orbital models from gravitational waveform data without environmental effects. The current project aims to advance GWI by incorporating environmental effects to discover new, detailed physical models. To this end, the team will develop specific models for dark matter halos and disc-embedded extreme mass ratio inspiral systems. The team will also focus on connecting their models to observations by interfacing with – and contributing to – open-source projects such as PyCBC and the Black Hole Perturbation Toolkit. This new approach could unlock the full potential of upcoming gravitational wave detectors, such as LISA, revealing precise information about binary black hole systems and their host environments. This award advances the goals of the NSF Windows on the Universe Big Idea through research in Multi-Messenger Astrophysics. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.