NeuronStar

Fast neural inference for neutron-star matter.

NeuronStar is the software of the paper Fast Bayesian Updating of the Neutron-Star Equation of State with Neural Posterior and Evidence Estimation by Prashant Thakur (Department of Physics, Yonsei University), arXiv:2610.05121. It provides amortized neural posterior estimation with importance-sampling correction (A-NET+IS), truncated sequential neural posterior estimation (TSNPE+MIS), and the Green-function Evidence Network (EN) for equation-of-state inference with nuclear, NICER, GW170817, and pQCD constraints.

Code: the full code and data, with step-by-step instructions to reproduce the paper, are on GitHub: https://github.com/prashantstar123/NeuronStar. This PyPI release holds the package name and does not yet contain the code, so please use the GitHub repository. Contributions (issues and pull requests) are welcome there.

Citation

@article{Thakur:2026smp,
    author = "Thakur, Prashant",
    title = "{Fast Bayesian Updating of the Neutron-Star Equation of State with Neural Posterior and Evidence Estimation}",
    eprint = "2610.05121",
    archivePrefix = "arXiv",
    primaryClass = "astro-ph.HE",
    month = "10",
    year = "2026"
}