Parameter inference of millilensed gravitational waves using neural spline flows

TL;DR

This paper introduces neural spline flows for efficient, accurate inference of lens parameters in millilensed gravitational waves, significantly reducing computation time compared to traditional Bayesian methods.

Contribution

The authors develop a neural spline flow-based method for rapid posterior inference of lens parameters in gravitational waves, achieving comparable accuracy with much faster processing.

Findings

NSFs achieve inference accuracy comparable to traditional Bayesian methods.

Inference time is reduced from about 3 days to 0.8 seconds.

The method generalizes well to source spin parameters.

Abstract

When gravitational waves (GWs) propagate near massive objects, they undergo gravitational lensing that imprints lens model dependent modulations on the waveform. This effect provides a powerful tool for cosmological and astrophysical studies. Due to the added parameters of lenses and the uncertainty of lens models, parameter inference for lensed GW events using traditional methods is extremely time-consuming, thus requiring more efficient parameter inference methods. In this work, we explore the use of neural spline flows (NSFs) for posterior inference of millilensed GWs, and successfully apply NSFs to the inference of 11-dimensional lens parameters. Our results demonstrate that compared with traditional methods like Bilby dynesty that rely on Bayesian inference, the NSF network we built not only achieves inference accuracy comparable to traditional methods for most parameters, but also can reduce the inference time from approximately 3 days to 0.8 s on average. Additionally, the network exhibits strong generalization for the spin parameters of GW sources. It is anticipated to become a powerful tool for future low-latency searches for lensed GW signals.