Benchmarking Exact, Gaussian Process-emulated, and SBI for Late-time Cosmology
- 1SR University
Abstract
Forthcoming cosmological surveys require inference pipelines that are both statistically reliable and computationally scalable. In this work, we perform a systematic comparison of three complementary inference strategies for late-time ΛCDM cosmology: exact Markov Chain Monte Carlo (MCMC), Gaussian process (GP)-assisted MCMC, and neural simulation-based inference (SBI). Using a common analysis framework based on cosmic chronometers (CC), DESI DR2 baryon acoustic oscillation measurements, and the PP Type Ia supernova compilation, we consider two dataset combinations of increasing complexity, namely, CC+DESI and CC+DESI+PP, under identical cosmological assumptions and priors. For CC+DESI, both GP emulation and SBI reproduce the exact posterior constraints on ( H ₀ , Ω m ,0 ) to better than 0.3 σ . For the more constraining CC+DESI+PP combination, modest method-dependent shifts emerge, reaching at most ∼1.5 σ in a single parameter. Despite these differences, all methods recover a nearly identical expansion history, with percent-level agreement across the full redshift range. From a computational perspective, GP emulation accelerates model evaluations but remains limited by MCMC sampling, whereas SBI achieves order-of-magnitude reductions in total runtime through amortized posterior learning. We further investigate the convergence of SBI as a function of simulation budget and identify the number of simulations required to obtain stable posterior constraints. Overall, our results demonstrate that accelerated inference techniques can deliver reliable cosmological constraints for realistic late-time datasets considered here at a fraction of the computational cost of conventional likelihood-based analyses.
Versions
Related
- Benchmarking Exact, GP-Emulated, and Simulation-Based Inference for Late-Time Cosmology
- Enhancing dark siren cosmology via Gaussian process reconstruction of incomplete galaxy catalogs
- Gaussian process inference of stochastic magneto-active dynamics and viscosity in Swift J1727.8-1613
- A Spatio-Temporal Self-Propagating Log-Gaussian Cox-Hawkes Process for Star Formation Modelling
- Hierarchical Gaussian-process reconstruction of late-time expansion with DESI BAO and compressed Planck priors
BibTeX
@article{mishra2026benchmarking,
title = {{Benchmarking Exact, Gaussian Process-emulated, and SBI for Late-time Cosmology}},
author = {Sai Swagat Mishra},
journal = {The Astrophysical Journal},
year = {2026},
month = oct,
doi = {10.3847/1538-4357/aea839},
url = {https://iopscience.iop.org/article/10.3847/1538-4357/aea839},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Benchmarking Exact, Gaussian Process-emulated, and SBI for Late-time Cosmology
AU - Sai Swagat Mishra
PY - 2026
DA - 2026/10/08
JO - The Astrophysical Journal
DO - 10.3847/1538-4357/aea839
UR - https://iopscience.iop.org/article/10.3847/1538-4357/aea839
AB - Forthcoming cosmological surveys require inference pipelines that are both statistically reliable and computationally scalable. In this work, we perform a systematic comparison of three complementary inference strategies for late-time ΛCDM cosmology: exact Markov Chain Monte Carlo (MCMC), Gaussian process (GP)-assisted MCMC, and neural simulation-based inference (SBI). Using a common analysis framework based on cosmic chronometers (CC), DESI DR2 baryon acoustic oscillation measurements, and the PP Type Ia supernova compilation, we consider two dataset combinations of increasing complexity, namely, CC+DESI and CC+DESI+PP, under identical cosmological assumptions and priors. For CC+DESI, both GP emulation and SBI reproduce the exact posterior constraints on ( H ₀ , Ω m ,0 ) to better than 0.3 σ . For the more constraining CC+DESI+PP combination, modest method-dependent shifts emerge, reaching at most ∼1.5 σ in a single parameter. Despite these differences, all methods recover a nearly identical expansion history, with percent-level agreement across the full redshift range. From a computational perspective, GP emulation accelerates model evaluations but remains limited by MCMC sampling, whereas SBI achieves order-of-magnitude reductions in total runtime through amortized posterior learning. We further investigate the convergence of SBI as a function of simulation budget and identify the number of simulations required to obtain stable posterior constraints. Overall, our results demonstrate that accelerated inference techniques can deliver reliable cosmological constraints for realistic late-time datasets considered here at a fraction of the computational cost of conventional likelihood-based analyses.
SN - 0004-637X
SN - 1538-4357
ER -