High-accuracy remnant kicks, spins, and masses from precessing binaries with NR surrogate modeling

Gravitational waves (GWs) carry energy, angular momentum, and linear momentum; the black hole remnant from a binary merger has its mass, spin, and recoil "kick" velocity determined by the emitted GWs. These quantities are astrophysically important for binary population synthesis, waveform modeling, and developing GW-based tests of GR. We present accurate fits for the remnant properties of generically precessing binary black holes, trained directly on large banks of numerical-relativity simulations, using modern machine-learning techniques. Our model is at least an order of magnitude more accurate than all previous fits, and changes the paradigm to avoid using manually-constructed fitting formulas. We make our models available in a fast, easy-to-use python package, surfinBH.