Discovery of Interpretable Surrogates via Agentic AI: Application to Gravitational Waves

TL;DR

This paper introduces GWAgent, an agentic AI workflow that constructs interpretable, accurate, and fast surrogate models for gravitational wave simulations, demonstrating improved performance and physical insight over existing methods.

Contribution

The work presents a novel LLM-based workflow for building interpretable surrogates directly from simulation data, incorporating physics-informed domain knowledge to enhance accuracy.

Findings

Achieved a median LIGO mismatch of 6.9×10^{-4} with GWAgent.

Provided an 8.4× speedup in waveform evaluation.

Successfully inferred eccentricity of GW200129 from data.

Abstract

Fast surrogate models for expensive simulations are now essential across the sciences, yet they typically operate as black boxes. We present \texttt{GWAgent}, a large language model (LLM)-based workflow that constructs interpretable analytic surrogates directly from simulation data. Surrogate modeling is well suited to agentic workflows because candidate models can be quantitatively validated against ground-truth simulations at each iteration. As a demonstration, we build a surrogate for gravitational waveforms from eccentric binary black hole mergers. We show that providing the agent with a physics-informed domain ansatz substantially improves output model accuracy. The resulting analytic surrogate attains a median Advanced LIGO mismatch of $6.9\times10^{-4}$ together with an $\sim 8.4\times$ speedup in waveform evaluation, surpassing both symbolic regression and conventional machine learning baselines. Beyond producing an accurate model, the workflow identifies compact physical structure from the learned representation. As an astrophysical application, we use \texttt{GWAgent} to analyze the eccentricity of GW200129 and infer $e_{20\mathrm{Hz}}=0.099^{+0.063}_{-0.044}$. These results show that validation-constrained agentic workflows can produce accurate, fast, and interpretable surrogates for scientific simulations and inference.