Machine Learning Techniques for Stellar Light Curve Classification
Problem
Predicting and classifying stellar properties from noisy, sparse real Kepler light-curve time-series data, a task previously studied mostly on simulated data.
Why it matters
Extracting physical stellar properties from real survey data extends machine learning's usefulness in astronomy.
What the AI did
Machine learning was applied to over 94 GB of Kepler light curves to classify 10 stellar properties, comparing representation learning (LSTM RNN) with feature engineering; the LSTM failed while feature engineering succeeded for classification and regression.
Results
Feature engineering achieved values for stellar density, stellar radius, and effective temperature; no numeric metrics stated in evidence.
Impact
Claimed as one of the first studies using real light-curve data for machine learning, with mixed success across approaches.
Record
| Venue | The Astronomical Journal |
|---|---|
| Published | 2018 |
| Work type | article |
| Authors | Trisha A. Hinners, Kevin Tat, Rachel Thorp |
| Citations | 55 (OpenAlex, at import time) |
| References | 34 |
| Model or method | long short-term memory recurrent neural network; feature engineering approaches |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3847/1538-3881/aac16d |
| Open access | bronze |
| Keywords | machine learning, Kepler light curves, LSTM, feature engineering, stellar properties, classification |
Importance rating
incremental / niche
Early ML application to real Kepler light curves; abstract claims foundation for future astrophysical analysis tools but no stated scale of scientific or practical impact.
More in Physics & Astronomy
| # | Paper | Year | Citations | Importance |
|---|---|---|---|---|
| 1 | Magnetic control of tokamak plasmas through deep reinforcement learning | 2022 | 739 |
7
|
| 2 | Solving the quantum many-body problem with artificial neural networks | 2017 | 2,205 |
7
|
| 3 | Learning high-accuracy error decoding for quantum processors | 2024 | 90 |
7
|
| 4 | Optimizing Quantum Error Correction Codes with Reinforcement Learning | 2019 | 158 |
7
|
| 5 | Machine-learning-assisted correction of correlated qubit errors in a topological code | 2018 | 141 |
7
|