E-INSPIRE - II. Finding relics from wide-sky multi-band surveys: A proof-of-concept machine learning regression algorithm
Abstract
In this second paper of the E-INSPIRE series, we train a machine-learning-based regression on ∼430 nearby (z<0.5) ultra-compact massive galaxies (UCMGs) with spectroscopically inferred kinematics, stellar population parameters and a measured ``degree of relicness'' (DoR). Our goal is to investigate how robustly the spectroscopically inferred DoR can be statistically reconstructed from observable galaxy properties, and to explore the potential applicability of this framework to future wide-area surveys. We test several regression algorithms finding that Support Vector Regression (SVR) provides the best performance. We explore multiple input feature configurations, from a minimal set including only age and metallicity to more comprehensive ones incorporating stellar population parameters, kinematics, structural properties, and the associated uncertainties. All tested models achieve similarly high performance on the training set (R2≥0.81), except for the minimal configuration (R2∼0.78). When evaluated on an independent INSPIRE sample of 52 UCMGs, the predictive power remains robust, although with increased model-to-model variation. The DoR distribution shows three regimes, with low (DoR<0.3) and high (DoR>0.6) values sparsely populated, leading to mild regression shrinkage toward intermediate values. However, this behaviour enables a conservative selection strategy: galaxies with predicted DoR≥0.6 are strongly biased toward genuine extreme relics, making them prime targets for follow-up observations. This proof-of-concept confirms that the spectroscopically inferred DoR is robustly connected to observable stellar population and kinematical properties, and provides a first step toward future relic-candidate selection strategies in large photometric and spectroscopic surveys.
Related
- Unsupervised selection and characterisation of little red dots in JWST surveys with manifold learning
- Finding Novel Precursors for Solar Wind Stream Interaction Regions With Interpretable Deep Learning
- Spectroscopy-informed Deep Learning Searches for Photometric Little Red Dots in JWST Surveys
- Unsupervised selection and characterisation of little red dots in JWST surveys with manifold learning
- Identifying Planetary Nebulae Candidates from IPHAS and VPHAS+ Surveys via YOLO Deep Learning Models
BibTeX
@misc{rosen2026einspire,
title = {{E-INSPIRE - II. Finding relics from wide-sky multi-band surveys: A proof-of-concept machine learning regression algorithm}},
author = {Charles Rosen and Chiara Spiniello and John Mills and Alexey Sergeyev and Vladyslav Khramtsov and Anna Ferré-Mateu and Johanna Hartke and Michalina Maksymowicz-Maciata and Malgorzata Siudek and Crescenzo Tortora},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.12608},
eprint = {2609.12608},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.12608},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - E-INSPIRE - II. Finding relics from wide-sky multi-band surveys: A proof-of-concept machine learning regression algorithm
AU - Charles Rosen
AU - Chiara Spiniello
AU - John Mills
AU - Alexey Sergeyev
AU - Vladyslav Khramtsov
AU - Anna Ferré-Mateu
AU - Johanna Hartke
AU - Michalina Maksymowicz-Maciata
AU - Malgorzata Siudek
AU - Crescenzo Tortora
PY - 2026
DA - 2026/09/11
PB - arXiv
DO - 10.48550/arxiv.2609.12608
UR - https://arxiv.org/abs/2609.12608
AB - In this second paper of the E-INSPIRE series, we train a machine-learning-based regression on ∼430 nearby (z<0.5) ultra-compact massive galaxies (UCMGs) with spectroscopically inferred kinematics, stellar population parameters and a measured ``degree of relicness'' (DoR). Our goal is to investigate how robustly the spectroscopically inferred DoR can be statistically reconstructed from observable galaxy properties, and to explore the potential applicability of this framework to future wide-area surveys. We test several regression algorithms finding that Support Vector Regression (SVR) provides the best performance. We explore multiple input feature configurations, from a minimal set including only age and metallicity to more comprehensive ones incorporating stellar population parameters, kinematics, structural properties, and the associated uncertainties. All tested models achieve similarly high performance on the training set (R^2≥0.81), except for the minimal configuration (R^2∼0.78). When evaluated on an independent INSPIRE sample of 52 UCMGs, the predictive power remains robust, although with increased model-to-model variation. The DoR distribution shows three regimes, with low (DoR<0.3) and high (DoR>0.6) values sparsely populated, leading to mild regression shrinkage toward intermediate values. However, this behaviour enables a conservative selection strategy: galaxies with predicted DoR≥0.6 are strongly biased toward genuine extreme relics, making them prime targets for follow-up observations. This proof-of-concept confirms that the spectroscopically inferred DoR is robustly connected to observable stellar population and kinematical properties, and provides a first step toward future relic-candidate selection strategies in large photometric and spectroscopic surveys.
ER -