AI and extreme scale computing to learn and infer the physics of higher order gravitational wave modes of quasi-circular, spinning, non-precessing binary black hole mergers

TL;DR

This paper develops AI models trained on 14 million waveforms to accurately and rapidly infer the physics of higher order gravitational wave modes in binary black hole mergers, outperforming traditional methods.

Contribution

The authors created AI surrogates trained on extensive waveform data that significantly improve inference accuracy and speed for gravitational wave parameters.

Findings

AI models outperform traditional inference methods in accuracy.

AI inference is three to four orders of magnitude faster.

Models trained on 14 million waveforms within 3.4 hours on supercomputers.

Abstract

We use artificial intelligence (AI) to learn and infer the physics of higher order gravitational wave modes of quasi-circular, spinning, non precessing binary black hole mergers. We trained AI models using 14 million waveforms, produced with the surrogate model NRHybSur3dq8, that include modes up to $\ell \leq 4$ and $(5,5)$, except for $(4,0)$ and $(4,1)$, that describe binaries with mass-ratios $q\leq8$, individual spins $s^z_{\{1,2\}}\in[-0.8, 0.8]$, and inclination angle $\theta\in[0,\pi]$.Our probabilistic AI surrogates can accurately constrain the mass-ratio, individual spins, effective spin, and inclination angle of numerical relativity waveforms that describe such signal manifold. We compared the predictions of our AI models with Gaussian process regression, random forest, k-nearest neighbors, and linear regression, and with traditional Bayesian inference methods through the PyCBC Inference toolkit, finding that AI outperforms all these approaches in terms of accuracy, and are between three to four orders of magnitude faster than traditional Bayesian inference methods. Our AI surrogates were trained within 3.4 hours using distributed training on 1,536 NVIDIA V100 GPUs in the Summit supercomputer.