Gamma-ray Bursts as Cosmological Probes Calibrated with Strongly Lensed Quasars and Neural Networks
Abstract
Gamma-ray bursts (GRBs) extend distance measurements to redshifts beyond those routinely probed by Type Ia supernovae, but their use in cosmology is limited by calibration circularity and possible redshift-dependent effects. We investigate strongly lensed quasars as external calibrators of the Amati and Combo GRB relations. We reconstruct the luminosity-distance relation using an artificial neural network trained on TDCOSMO strong-lensing distances, with the absolute scale inherited from the lensing-only calibration. To avoid extrapolation, only GRBs within the lens-supported redshift interval are used for calibration. We then infer cosmological parameters through a hierarchical Bayesian analysis of flat ΛCDM, with w0waCDM examined as a consistency test. For flat ΛCDM, Amati yields H0=78.26+17.87−16.30 km s−1 Mpc−1 and Ωm=0.670+0.223−0.276, while Combo yields H0=78.71+13.83−11.29 km s−1 Mpc−1 and Ωm=0.635+0.235−0.242. Amati is relatively insensitive to the phenomenological redshift term, whereas Combo shows a stronger dependence. Injection--recovery tests indicate that an unmodelled positive Combo trend can bias H0 and Ωm upward. Strong-lens calibration therefore offers a route to extending the GRB Hubble diagram, but the inferred H0 is not an independent GRB measurement: it retains the TDCOSMO distance scale. The precision is limited by GRB scatter, redshift-dependent systematics, and lens-modelling uncertainties.
Related
- The host halo masses of AGNs and quasars at z ∼ 3 − 7 with TNG-Cluster, FLAMINGO and other cosmological galaxy simulations
- Ray-tracing Fast Radio Bursts Through IllustrisTNG: Cosmological Dispersion Measures from Redshift 0 to 5.5
- Assessing the consistency of the CPL dark-energy model with cosmological data: Simulation-calibrated neural posterior disagreement
- Strongly mixed cosmological collider at unequal sound speeds
- Model-independent calibration of gamma-ray bursts with Bayesian neural networks
BibTeX
@misc{hong2026gammaray,
title = {{Gamma-ray Bursts as Cosmological Probes Calibrated with Strongly Lensed Quasars and Neural Networks}},
author = {Wei Hong and Luca Izzo and Massimo Della Valle and Orlando Luongo and Marco Muccino and Tong-Jie Zhang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.27009},
eprint = {2610.27009},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.27009},
}
RIS
TY - UNPB
TI - Gamma-ray Bursts as Cosmological Probes Calibrated with Strongly Lensed Quasars and Neural Networks
AU - Wei Hong
AU - Luca Izzo
AU - Massimo Della Valle
AU - Orlando Luongo
AU - Marco Muccino
AU - Tong-Jie Zhang
PY - 2026
DA - 2026/10/19
PB - arXiv
DO - 10.48550/arxiv.2610.27009
UR - https://arxiv.org/abs/2610.27009
AB - Gamma-ray bursts (GRBs) extend distance measurements to redshifts beyond those routinely probed by Type Ia supernovae, but their use in cosmology is limited by calibration circularity and possible redshift-dependent effects. We investigate strongly lensed quasars as external calibrators of the Amati and Combo GRB relations. We reconstruct the luminosity-distance relation using an artificial neural network trained on TDCOSMO strong-lensing distances, with the absolute scale inherited from the lensing-only calibration. To avoid extrapolation, only GRBs within the lens-supported redshift interval are used for calibration. We then infer cosmological parameters through a hierarchical Bayesian analysis of flat ΛCDM, with w_0w_aCDM examined as a consistency test. For flat ΛCDM, Amati yields H_0=78.26^+17.87_−16.30 km s^−1 Mpc^−1 and Ω_m=0.670^+0.223_−0.276, while Combo yields H_0=78.71^+13.83_−11.29 km s^−1 Mpc^−1 and Ω_m=0.635^+0.235_−0.242. Amati is relatively insensitive to the phenomenological redshift term, whereas Combo shows a stronger dependence. Injection--recovery tests indicate that an unmodelled positive Combo trend can bias H_0 and Ω_m upward. Strong-lens calibration therefore offers a route to extending the GRB Hubble diagram, but the inferred H_0 is not an independent GRB measurement: it retains the TDCOSMO distance scale. The precision is limited by GRB scatter, redshift-dependent systematics, and lens-modelling uncertainties.
ER -