Accelerated Bayesian parameter estimation and model selection for gravitational waves with normalizing flows

TL;DR

This paper introduces a high-performance, GPU-accelerated Bayesian inference pipeline using normalizing flows for gravitational wave data, significantly reducing computation time while maintaining accuracy.

Contribution

It develops an efficient, GPU-based Bayesian inference method with normalizing flows, integrating existing tools for faster gravitational wave parameter estimation and model selection.

Findings

GPU-based evidence estimates match traditional methods

Achieves 5-15x speedup in inference tasks

Open-source implementation ensures reproducibility

Abstract

We present an accelerated pipeline, based on high-performance computing techniques and normalizing flows, for joint Bayesian parameter estimation and model selection and demonstrate its efficiency in gravitational wave astrophysics. We integrate the Jim inference toolkit, a normalizing flow-enhanced Markov chain Monte Carlo (MCMC) sampler, with the learned harmonic mean estimator. Our Bayesian evidence estimates run on $1$ GPU are consistent with traditional nested sampling techniques run on $16$ CPU cores, while reducing the computation time by factors of $5\times$ and $15\times$ for $4$-dimensional and $11$-dimensional gravitational wave inference problems, respectively. Our code is available in well-tested and thoroughly documented open-source packages, ensuring accessibility and reproducibility for the wider research community.