Explaining Neural Networks on the Sky: Machine Learning Interpretability for CMB Maps https://github.com/skyexplain
G. Cañas-Herrera · 2026
Read Explaining Neural Networks on the Sky: Machine Learning Interpretability for CMB Maps https://github.com/skyexplain by G. Cañas-Herrera (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-2604-05290.txt.
Description
Ocampo and Cañas-Herrera present a framework for cosmological selection using neural networks trained on microwave background maps. SHAP maps the network's choice process for interpretability. A galactic mask as a control yields maps leaving out that region, showing sky-driven classification. These maps verify the architecture distinguishes Lambda-CDM from a feature model via variance across regions. The work describes CMB maps and a hybrid architecture combining PCA and neural networks for classification. It introduces SkyExplain, giving pipeline access for runs, training, and interpretability.
Catalog record
| CREATOR | G. Cañas-Herrera |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46665 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2604.05290