Is the ‘Known’ Enough? An Integrated Machine Learning Framework for Eclipsing Binary Classification and Parameter Estimation Based on Well-Characterized Systems
Burak Ulaş · 2026
Read Is the ‘Known’ Enough? An Integrated Machine Learning Framework for Eclipsing Binary Classification and Parameter Estimation Based on Well-Characterized Systems by Burak Ulaş (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2604-19690.txt.
Description
Ulaş builds a multi-task machine learning tool that classifies eclipsing binaries by shape and estimates their physical traits from light curves at once. Random Forest and XGBoost models train on 845 of 995 well-studied systems across three shape groups, using 51 tailored features drawn from each folded light curve. 15 percent of systems were held back for testing before any training. On that held-out set, XGBoost scores strongly across the board, with the mass ratio weakest at 0.77 among the five measures.
Catalog record
| CREATOR | Burak Ulaş |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 88800 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.19690