Speaker
Description
Gravitational waves from compact binaries provide excellent opportunities for testing general relativity (GR) in the strong- and dynamical-field regime. So far, tests using inspiral signals have employed certain variants of the parametrized post-Einsteinian (ppE) framework to model deviations from GR. However, this approach has several limitations: the parameters introduced do not fully encompass all potential deviations from GR, and the estimation of these parameters is computationally expensive. To address the above issues, a neural post-Einsteinian (npE) framework has been developed by extending the ppE parametrization with a generative neural network, and the performance of the npE framework has been demonstrated using synthetic signals. In this talk, I will present recent results from applying the npE framework to the real data in the third Gravitational-Wave Transient Catalog. With a broader range of non-GR deviations incorporated in the npE parameter space, the test-of-GR conclusions can be drawn with greater robustness.
