Advisor: Hang Yu
Binary neutron stars (BNSs) are crucial sources for gravitational wave observations, as they offer a simultaneous probe of the spacetime at extreme curvature and behavior of matter at extreme density. To detect a BNS, theoretical waveform templates are required, which are computationally expensive to generate, typically taking minutes per template evaluation. To analyze a BNS event, however, millions of template evaluations are necessary. In this project, we will explore using machine learning techniques to generate efficient surrogate models that preserve the accuracy while being evaluated in under a second.