Uncertainty-Aware Deep Learning for the Lyα Forest: CNN-Based Absorber Detection and Characterization
4 affiliations
- 1Central University of Himachal Pradesh Department of Physics and Astronomical Science, , Dharamshala, 176215 ,
- 2Indian Institute of Technology Tirupati Department of Physics, , Tirupati, Andhra Pradesh 517619 ,
- 3Kapteyn Astronomical Institute, University of Groningen , Landleven 12 (Kapteynborg, 5419), 9747 AD Groningen ,
- 4Astrophysics and Space Engineering, Indian Institute of Technology Indore Department of Astronomy, , Simrol, MP 453552 ,
Abstract
The Lyα forest is a powerful probe of the intergalactic medium and small-scale matter distribution, but deriving absorber properties traditionally requires computationally expensive Voigt-profile fitting. We present a convolutional neural network (CNN) that identifies and characterizes H i Lyα absorbers directly from quasar spectra. The model is trained on synthetic spectra generated from the IllustrisTNG simulation and fitted with the viper Voigt-profile fitting code to provide training labels. The network simultaneously predicts absorber presence, column density (NHI), Doppler parameter (bHI), and line centroid. On simulated spectra, the CNN reproduces the viper reference catalogue with an F1 score of ∼0.8, with mean absolute errors of ∼0.18 in log NHI and ∼0.10 in log bHI. It reproduces the H i column density distribution function (CDDF) and the bHI–NHI relation obtained from viper, recovering similar CDDF slopes and a lower-envelope relation with an RMS difference of only 0.36 km s−1. Applied to high-resolution UVES spectra, performance decreases to an F1 score of ∼0.5, with mean absolute errors of ∼0.35 in log NHI and ∼0.22 in log bHI. Latent-space analysis reveals a significant domain shift between the simulated and observational spectra, contributing to the reduced performance. Nevertheless, the CNN reproduces the observed CDDF and bHI–NHI distributions obtained from the viper reference catalogue, yielding CDDF slopes consistent with viper and a lower-envelope RMS difference of 3.20 km s−1 relative to viper. Monte Carlo dropout is implemented during inference to quantify predictive uncertainties. Together with its computational efficiency, the method provides a scalable and uncertainty-aware framework for Lyα forest analysis in upcoming spectroscopic surveys.
Related
- Uncertainty-Aware Tidal Disruption Event Classification : A Host-Agnostic Probabilistic Random Forest Approach
- Uncertainty-Aware Tidal Disruption Event Classification : A Host-Agnostic Probabilistic Random Forest Approach
- Where the Forest Goes Dark: Halo-centered Characterization of High Column Density Systems with IllustrisTNG
- Interpretable human-label-free deep learning for real-bogus classification with uncertainty quantification
- Super-Resolution of Solar Magnetograms via Adaptive Stratified Ensemble Learning with Uncertainty Estimation
BibTeX
@article{sharma2026uncertaintyaware,
title = {{Uncertainty-Aware Deep Learning for the Lyα Forest: CNN-Based Absorber Detection and Characterization}},
author = {Paryag Sharma and Vikram Khaire and Ting-Yun Cheng and Hum Chand and Prakash Gaikwad},
journal = {Monthly Notices of the Royal Astronomical Society},
year = {2026},
month = oct,
doi = {10.1093/mnras/stag2027},
url = {https://academic.oup.com/mnras/advance-article/doi/10.1093/mnras/stag2027/8994288},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Uncertainty-Aware Deep Learning for the Lyα Forest: CNN-Based Absorber Detection and Characterization
AU - Paryag Sharma
AU - Vikram Khaire
AU - Ting-Yun Cheng
AU - Hum Chand
AU - Prakash Gaikwad
PY - 2026
DA - 2026/10/23
JO - Monthly Notices of the Royal Astronomical Society
DO - 10.1093/mnras/stag2027
UR - https://academic.oup.com/mnras/advance-article/doi/10.1093/mnras/stag2027/8994288
AB - The Lyα forest is a powerful probe of the intergalactic medium and small-scale matter distribution, but deriving absorber properties traditionally requires computationally expensive Voigt-profile fitting. We present a convolutional neural network (CNN) that identifies and characterizes H i Lyα absorbers directly from quasar spectra. The model is trained on synthetic spectra generated from the IllustrisTNG simulation and fitted with the viper Voigt-profile fitting code to provide training labels. The network simultaneously predicts absorber presence, column density (NHI), Doppler parameter (bHI), and line centroid. On simulated spectra, the CNN reproduces the viper reference catalogue with an F1 score of ∼0.8, with mean absolute errors of ∼0.18 in log NHI and ∼0.10 in log bHI. It reproduces the H i column density distribution function (CDDF) and the bHI–NHI relation obtained from viper, recovering similar CDDF slopes and a lower-envelope relation with an RMS difference of only 0.36 km s−1. Applied to high-resolution UVES spectra, performance decreases to an F1 score of ∼0.5, with mean absolute errors of ∼0.35 in log NHI and ∼0.22 in log bHI. Latent-space analysis reveals a significant domain shift between the simulated and observational spectra, contributing to the reduced performance. Nevertheless, the CNN reproduces the observed CDDF and bHI–NHI distributions obtained from the viper reference catalogue, yielding CDDF slopes consistent with viper and a lower-envelope RMS difference of 3.20 km s−1 relative to viper. Monte Carlo dropout is implemented during inference to quantify predictive uncertainties. Together with its computational efficiency, the method provides a scalable and uncertainty-aware framework for Lyα forest analysis in upcoming spectroscopic surveys.
SN - 0035-8711
SN - 1365-2966
ER -