The Sequential Monte Carlo goes NUTS: Boosting Gravitational-Wave Inference

TL;DR

SHARPy is a new Bayesian inference framework that combines SMC and NUTS algorithms, leveraging GPU acceleration to perform rapid and accurate gravitational-wave parameter estimation and evidence calculation.

Contribution

It introduces SHARPy, integrating SMC with NUTS and geometric information, enabling fast, GPU-accelerated gravitational-wave inference with reliable evidence estimation.

Findings

Performs inference in around ten minutes on GPU

Produces posterior samples and evidence estimates consistent with Nested Sampling

Sets a new milestone in likelihood-based GW inference

Abstract

Sequential Monte Carlo (SMC) methods have recently been applied to gravitational-wave inference as a powerful alternative to standard sampling techniques, such as Nested Sampling. At the same time, gradient-based Markov Chain Monte Carlo algorithms, most notably the No-U-Turn Sampler (NUTS), provide an efficient way to explore high-dimensional parameter spaces. In this work we present SHARPy, a Bayesian inference framework that combines the parallelism and evidence-estimation capabilities of SMC with the state-of-the-art sampling performance of NUTS. Moreover, SHARPy exploits the local geometric structure of the posterior to further improve efficiency. Built on JAX and accelerated on GPUs, SHARPy performs gravitational-wave inference on binary black-hole events in around ten minutes, yielding posterior samples and Bayesian evidence estimates that are consistent with those obtained through Nested Sampling. This work sets a new milestone in GW inference with likelihood-based methods and paves the way for model comparison tasks to be accomplished in minutes.