# Sound-Horizon-Independent Test of Cosmic Distance Duality Relation Using Artificial Neural Networks and Gaussian Processes

Preprint · Bo-Hao Jiang, Guo-Jian Wang, Tao Yang

- Archive id: arxiv:2609.24203
- Published: 2026-09-21
- arXiv: https://arxiv.org/abs/2609.24203
- Source URL: https://arxiv.org/abs/2609.24203
- PDF: https://arxiv.org/pdf/2609.24203
- Provenance: arxiv

## Abstract

The cosmic distance duality relation (CDDR), $D\_L(z)(1+z)^{-2}/D\_A(z) \\equiv 1$, is a fundamental relation in modern cosmology linking luminosity distance $D\_L$ and angular diameter distance $D\_A$. We define $η(z) \\equiv D\_L(z)(1+z)^{-2}/D\_A(z)$ and reconstruct the relevant quantities over $0 \< z \< 2.5$ using two independent non-parametric methods: Gaussian processes (GP) and artificial neural networks (ANN). Specifically, we reconstruct $D\_L$ from three different Type Ia supernovae (SNe$~$Ia) compilations (PantheonPlus, Union3, and DES-Dovekie), the baryon acoustic oscillation distance ratio $D\_M/D\_H$ from SDSS and DESI, and the Hubble function $H(z)$ from cosmic chronometer (CC) data. Our CDDR test is independent of both the cosmological model and the sound horizon $r\_d$ calibration, as $D\_M/D\_H$ naturally eliminates $r\_d$. We render our GP reconstructions insensitive to mean-function and kernel choices via full Bayesian marginalization over hyperparameters, with ANN as a cross-check$-$revealing the impact of data sparsity on the latter. Our results show no significant deviation from the CDDR exceeding $2σ$, and for the CC+BAO+DES-Dovekie combination it remains within $1σ$ over most of the redshift range. With $r\_d$ eliminated, varying the only remaining external parameter $M\_B$ systematically shifts $η(z)$ through the $M\_B$-$H\_0$ degeneracy, indicating that the observed $η\\neq1$ signal is closely tied to $H\_0$ tension among datasets and is likely driven by $M\_B$ systematics rather than a true CDDR violation. Future high-quality CC, BAO, and SNe$~$Ia data$-$with either improved $M\_B$ calibration or $M\_B$-free distance measurements$-$will be essential for extending this calibration-free approach.

_Abstract reproduced under CC0-1.0 (arXiv descriptive metadata)._

---
Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.