Description
This work explores the use of machine learning to emulate stellar evolution codes, which are used to predict complete stellar tracks and detailed global and internal properties throughout a star’s lifetime. Although these codes are essential tools in astrophysical research, they are computationally expensive and require significant manual effort. Alternative approaches, such as analytic fitting or interpolation, are poorly automated and introduce substantial systematic errors. This study focuses on the Red Giant Branch (RGB) phase, a particularly complex and time-consuming stage of stellar evolution to model. We evaluate classical machine learning and deep learning models trained on pre-computed stellar track grids to efficiently reproduce and time-skip the RGB phase. This approach aims to achieve improved computational efficiency compared to existing methods and provide physical insight into the key parameters governing stellar evolution.
| Field of Research/Work | Cosmology, Astrophysics, and Gravitation |
|---|
