ASTRAFier: A Novel and Scalable Transformer-based Stellar Variability Classifier

TL;DR

This paper introduces ASTRAFier, a scalable Transformer-based model that classifies stellar variability directly from light curves, achieving high accuracy on Kepler and TESS data, and applied to millions of light curves for a new catalog.

Contribution

The paper presents a novel Transformer-based model combining BiLSTM and CNN for direct light curve classification, eliminating feature engineering and enabling large-scale deployment.

Findings

Achieved 94.26% accuracy on Kepler data.

Achieved 88.22% accuracy on TESS data.

Deployed on 2.8 million TESS light curves, releasing a variability catalog.

Abstract

Photometric missions such as Kepler and TESS have generated millions of light curves covering almost the entire sky, offering unprecedented opportunities to study stellar variability and advance our understanding of the Universe. In this data-rich environment, machine learning has emerged as a powerful tool to efficiently and accurately process and classify light curves according to their type of stellar variability. In this work, we introduce ASTRAFier: a novel Transformer-based model for variability classification that integrates Bidirectional Long Short-Term Memory (BiLSTM) and Convolutional Neural Networks (CNNs). The model operates directly on time series without requiring feature engineering, creating an easy-to-maintain and efficient end-to-end classification framework. We train and validate our model using both Kepler and TESS light curves and, respectively, achieve a classification accuracy of $94.26\%$ on Kepler and $88.22\%$ on TESS. We demonstrate scalability by deploying our model on $\sim 2.8$ million TESS light curves from sectors 14, 15, and 26 (Kepler Field-of-View) delivered by MIT's Quick-look Pipeline (QLP) and release the resulting stellar variability catalog.