Estimating redshift distributions of Fermi blazars with multiple machine-learning models

Abstract

Approximately half of the Fermi-LAT-detected blazars lack spectroscopic redshift measurements, which limits population studies and investigations of the cosmological evolution of the gamma-ray blazar population. We aim to develop a multi-model ensemble framework to systematically evaluate the performance of machine-learning models for blazar redshift estimation and to provide predictive redshift probability distributions with quantified uncertainties for all blazars without measured redshifts. We construct an ensemble of 35 machine-learning regression models in the 1/(1+z) space. Predictions from individual models are combined to build non-parametric redshift probability density functions (PDFs) for each source, allowing an explicit characterization of predictive uncertainty. The source-level PDFs are subsequently stacked to derive population-level redshift distributions. Across the diverse models within our framework, we find comparable predictive performance, indicating a feature-limited performance plateau. We compile a catalog of predictive redshift distributions for 1,590 blazars without measured redshifts. The stacked PDFs recover the characteristic double-peaked redshift structure of blazars without using source class labels. For blazar candidates of uncertain type, the predicted redshift distribution exhibits a bimodal structure, suggesting a slightly larger contribution from FSRQ-like sources. Our multi-model statistical framework achieves competitive predictive performance while retaining faint sources. We provide predicted redshift distributions with quantified uncertainties for 1,590 blazars without measured redshifts, offering improved support for population and evolutionary studies. More broadly, this approach highlights the practical value of uncertainty-aware multi-model regression ensembling in astronomical data analysis.

The LGRBs Redshift and Jet Opening Angle Distributions and the Hubble Constant H₀
Truong Le, Joseph Chromicz
arXiv13 Sep 2026PreprintCC BY 4.0Full textHigh-energy astrophysicsarXiv
Anisotropic redshift distributions in photometric galaxy clustering and their cosmological impact
Zekang Zhang, Yunhe Wang, Daniel Gruen +4
arXiv14 Sep 2026PreprintarXiv licenceFull textCosmology GalaxiesarXiv
Classification of Fermi Blazars of Uncertain Type Using Positive-Unlabelled Machine Learning
Sikandar Akbar, Zahir Shah, Zahoor Malik +1
arXiv26 Oct 2026PreprintarXiv licenceFull textHigh-energy astrophysicsarXiv
A Stellar Role Reversal: Multiple Features in the Mass and Mass Ratio Distributions of Merging Binary Black Holes from Stable Mass Transfer
Gina Chen, Katelyn Breivik, Lieke A. C. van Son
arXiv24 Jun 2026PreprintCC BY 4.0Full textHigh-energy astrophysics Solar & stellar physicsarXiv
Combining Multiple Incomplete Astronomical Surveys to Estimate Galaxy Flux Distributions
Maximilian Autenrieth, David A. van Dyk, Vinay Kashyap +3
arXiv27 Oct 2026PreprintarXiv licenceFull textGalaxies Instrumentation & methodsarXiv

Cite

BibTeX RIS

BibTeX

@misc{zhu2026estimating,
  title = {{Estimating redshift distributions of Fermi blazars with multiple machine-learning models}},
  author = {Kerui Zhu and Jiaming Chen and Yonggang Zheng and Ruixin Zhou and Shan Chang and Li Zhang and Niansheng Tang},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.27397},
  eprint = {2609.27397},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.27397},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Estimating redshift distributions of Fermi blazars with multiple machine-learning models
AU  - Kerui Zhu
AU  - Jiaming Chen
AU  - Yonggang Zheng
AU  - Ruixin Zhou
AU  - Shan Chang
AU  - Li Zhang
AU  - Niansheng Tang
PY  - 2026
DA  - 2026/09/23
PB  - arXiv
DO  - 10.48550/arxiv.2609.27397
UR  - https://arxiv.org/abs/2609.27397
AB  - Approximately half of the Fermi-LAT-detected blazars lack spectroscopic redshift measurements, which limits population studies and investigations of the cosmological evolution of the gamma-ray blazar population. We aim to develop a multi-model ensemble framework to systematically evaluate the performance of machine-learning models for blazar redshift estimation and to provide predictive redshift probability distributions with quantified uncertainties for all blazars without measured redshifts. We construct an ensemble of 35 machine-learning regression models in the 1/(1+z) space. Predictions from individual models are combined to build non-parametric redshift probability density functions (PDFs) for each source, allowing an explicit characterization of predictive uncertainty. The source-level PDFs are subsequently stacked to derive population-level redshift distributions. Across the diverse models within our framework, we find comparable predictive performance, indicating a feature-limited performance plateau. We compile a catalog of predictive redshift distributions for 1,590 blazars without measured redshifts. The stacked PDFs recover the characteristic double-peaked redshift structure of blazars without using source class labels. For blazar candidates of uncertain type, the predicted redshift distribution exhibits a bimodal structure, suggesting a slightly larger contribution from FSRQ-like sources. Our multi-model statistical framework achieves competitive predictive performance while retaining faint sources. We provide predicted redshift distributions with quantified uncertainties for 1,590 blazars without measured redshifts, offering improved support for population and evolutionary studies. More broadly, this approach highlights the practical value of uncertainty-aware multi-model regression ensembling in astronomical data analysis.
ER  -

Abstract

Related

Cite