Replacing Gaussian Processes with Neural Networks in Pulsar Timing Array Inference of the Gravitational-Wave Background

TL;DR

This paper demonstrates that neural networks can replace Gaussian processes in pulsar timing array gravitational-wave inference, reducing computational costs while maintaining accurate posterior results.

Contribution

It introduces a neural network-based approach that replaces Gaussian-process interpolators in gravitational-wave background inference, improving efficiency.

Findings

Neural networks recover consistent posteriors compared to Gaussian processes.

Training and MCMC runtimes are significantly reduced with neural networks.

Largest efficiency gains observed for more computationally demanding models.

Abstract

Bayesian inference of nanohertz gravitational-wave background models in pulsar timing array analyses often relies on Gaussian-process interpolators to avoid repeated, computationally expensive strain-spectrum calculations. However, Gaussian-process training becomes a bottleneck for large training sets. We test whether probabilistic neural networks can replace Gaussian processes in this role for both a self-interacting dark matter model and a phenomenological environmental model. We find that neural networks recover consistent posteriors while significantly reducing both training and Markov chain Monte Carlo runtime, with the largest gains for the more computationally demanding model.