Low-latency parameter inference enabled by a Gaussian likelihood approximation for RIFT

TL;DR

This paper introduces a modification to the RIFT pipeline that significantly reduces latency in gravitational wave parameter inference, facilitating rapid follow-up observations of binary mergers.

Contribution

The authors propose a Gaussian likelihood approximation within RIFT to achieve low-latency inference for nonprecessing gravitational wave sources.

Findings

Achieved near real-time inference latency.

Validated approach on nonprecessing sources.

Potential for improved multimessenger follow-up timing.

Abstract

Rapid identification, characterization, and localization of gravitational waves from binary compact object mergers can enable well-informed follow-on multimessenger observations. In this work, we investigate a small modification to the RIFT parameter inference pipeline to enable extremely low-latency inference, tested here for nonprecessing sources.