Application of deep learning: A study of dark energy models in cosmology
Journal of Advanced Science and Research · 24.12.2025 · 1 (2) · 1-17
DOI: 10.65525/jasr.v1i2.15 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Advanced Science and Research, online erschienen am 24.12.2025. Verfasst von Gopal Sardar. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
In the present work we use deep learning tool to generate the mock data. In particular, the combination of long-short-term-memory (LSTM) and Bayesian Neural Networks (BNN) architecture is used to generate mock data. In order to train the deep learning model, we consider Pantheon + data which is consist of the distance modulus corresponding to the redshift parameter spanned over the range (1701 data points) and we generate mock data upto =4. We analyze three cosmological models namely, (i) , (ii) (CPL), (iii) (modified CPL model) considering Pantheon, Pantheon+ and generated mock data in addition with BAO data. We find that there is a small deviation of best-fit values of the model parameters if we consider Pantheon+ instead of Pantheon. However, there is no significant deviation of the model parameters considering upto z=3 (3096 data points), upto z=4 (4096 data points) and Panthenon+ data. Finally, we compare the models using Bayesian Evidence and find that model is the best model compared to the other models.
Zitieren
APA
Sardar, G. (2025). Application of deep learning: A study of dark energy models in cosmology. Journal of Advanced Science and Research, 1(2), 1-17. https://doi.org/10.65525/jasr.v1i2.15
Vancouver
Sardar G. Application of deep learning: A study of dark energy models in cosmology. Journal of Advanced Science and Research. 2025;1(2):1-17. doi:10.65525/jasr.v1i2.15
BibTeX
@article{Sardar2025,
author = {Sardar, Gopal},
title = {Application of deep learning: A study of dark energy models in cosmology},
journal = {Journal of Advanced Science and Research},
year = {2025},
volume = {1},
number = {2},
pages = {1--17},
doi = {10.65525/jasr.v1i2.15}
}
EUVIMED