Data-Driven Generation of Neutron Star Equations of State Using Variational Autoencoders

2026-01-29

AI Summary

This work addresses the challenge of efficiently generating neutron star equations of state (EOS) that satisfy astrophysical observational constraints. It proposes a structured variational autoencoder (VAE) that, for the first time, explicitly embeds supervised observables—such as maximum mass and canonical radius—into the latent space, alongside automatically learned latent variables. This framework enables data-driven EOS generation under physical constraints and facilitates integration with multi-messenger observations, including LIGO/Virgo gravitational-wave signals and pulsar mass–radius measurements. Evaluated on a Skyrme EOS dataset, the model achieves high-fidelity EOS reconstruction using only two supervised quantities and a single latent variable, yielding mean absolute percentage errors of approximately 0.15% for both maximum mass and canonical radius, thereby significantly enhancing both generative efficiency and physical consistency.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Generation

Application Category

• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

We develop a machine learning model based on a structured variational autoencoder (VAE) framework to reconstruct and generate neutron star (NS) equations of state (EOS). The VAE consists of an encoder network that maps high-dimensional EOS data into a lower-dimensional latent space and a decoder network that reconstructs the full EOS from the latent representation. The latent space includes supervised NS observables derived from the training EOS data, as well as latent random variables corresponding to additional unspecified EOS features learned automatically. Sampling the latent space enables the generation of new, causal, and stable EOS models that satisfy astronomical constraints on the supervised NS observables, while allowing Bayesian inference of the EOS incorporating additional multimessenger data, including gravitational waves from LIGO/Virgo and mass and radius measurements of pulsars. Based on a VAE trained on a Skyrme EOS dataset, we find that a latent space with two supervised NS observables, the maximum mass $(M_{\max})$ and the canonical radius $(R_{1.4})$, together with one latent random variable controlling the EOS near the crust--core transition, can already reconstruct Skyrme EOSs with high fidelity, achieving mean absolute percentage errors of approximately $(0.15\%)$ for $(M_{\max})$ and $(R_{1.4})$ derived from the decoder-reconstructed EOS.

Problem

• neutron star
• equation of state
• data-driven generation
• multimessenger constraints
• physical consistency

Innovation

• variational autoencoder
• neutron star equation of state
• latent space
• multimessenger constraints
• Bayesian inference