Key points are not available for this paper at this time.
We propose a new framework to predict stellar properties from light curves. We analyze the light-curve data from the Kepler space mission and develop a novel tool for deriving the stellar rotation periods for main-sequence stars. Using this tool, we provide the largest (108785 stars) and most accurate (an average error of 1. 6 Days) sample of stellar rotations to date. Our model, LightPred, is a novel deep-learning model designed to extract stellar rotation periods from light curves. The model utilizes a dual-branch architecture combining Long Short-Term Memory (LSTM) and Transformer components to capture both temporal and global features within the data. We train LightPred on a dataset of simulated light curves generated using a realistic spot model and enhance its performance through self-supervised contrastive pre-training on Kepler light curves. Our evaluation demonstrates that LightPred outperforms classical methods like the Autocorrelation Function (ACF) in terms of accuracy and robustness. We apply LightPred to the Kepler dataset, generating the largest catalog to date of stellar rotation periods for main-sequence stars. Our analysis reveals a systematic shift towards shorter periods compared to previous studies, suggesting a potential revision of stellar age estimates. We also investigate the impact of stellar activity on period determination and find evidence for a distinct period-activity relation. Additionally, we confirm tidal synchronization in eclipsing binaries with orbital periods shorter than 10 days. Our findings highlight the potential of deep learning in extracting fundamental stellar properties from light curves, opening new avenues for understanding stellar evolution and population demographics.
Kamai et al. (Tue,) studied this question.