ATAT: Astronomical Transformer for time series and Tabular data

¹ Department of Computer Science, Universidad de Concepción, Concepción, Chile
² Center for Data and Artificial Intelligence, Universidad de Concepción, Edmundo Larenas 310, Concepción, Chile
³ Millennium Institute of Astrophysics (MAS), Nuncio Monseñor Sotero Sanz 100, Of. 104, Providencia, Santiago, Chile
⁴ Center for Mathematical Modeling (CMM), Universidad de Chile, Beauchef 851, Santiago 8320000, Chile
⁵ Department of Electrical Engineering, Universidad de Chile, Av. Tupper 2007, Santiago 8320000, Chile
⁶ Data Observatory Foundation, Eliodoro Yáñez 2990, oficina A5, Santiago, Chile
⁷ Data and Artificial Intelligence Initiative (ID&IA), University of Chile, Santiago, Chile
⁸ Institute of Astronomy (IvS), Department of Physics and Astronomy, KU Leuven, Celestijnenlaan 200D, 3001 Leuven, Belgium
⁹ Instituto de Informática, Facultad de Ciencias de la Ingeniería, Universidad Austral de Chile, General Lagos 2086, Valdivia, Chile
¹⁰ European Southern Observatory, Karl-Schwarzschild-Strasse 2, 85748 Garching bei München, Germany
¹¹ Instituto de Física y Astronomía, Universidad de Valparaíso, Av. Gran Bretaña 1111, Playa Ancha, Casilla 5030, Chile
¹² Instituto de Astrofísica, Pontificia Universidad Católica de Chile, Av. Vicuña Mackenna 4860, 7820436 Macul, Santiago, Chile
¹³ Centro de Astroingeniería, Pontificia Universidad Católica de Chile, Av. Vicuña Mackenna 4860, 7820436 Macul, Santiago, Chile

Received: 2 February 2024
Accepted: 9 June 2024

Abstract

Context. The advent of next-generation survey instruments, such as the Vera C. Rubin Observatory and its Legacy Survey of Space and Time (LSST), is opening a window for new research in time-domain astronomy. The Extended LSST Astronomical Time-Series Classification Challenge (ELAsTiCC) was created to test the capacity of brokers to deal with a simulated LSST stream.

Aims. Our aim is to develop a next-generation model for the classification of variable astronomical objects. We describe ATAT, the Astronomical Transformer for time series And Tabular data, a classification model conceived by the ALeRCE alert broker to classify light curves from next-generation alert streams. ATAT was tested in production during the first round of the ELAsTiCC campaigns.

Methods. ATAT consists of two transformer models that encode light curves and features using novel time modulation and quantile feature tokenizer mechanisms, respectively. ATAT was trained on different combinations of light curves, metadata, and features calculated over the light curves. We compare ATAT against the current ALeRCE classifier, a balanced hierarchical random forest (BHRF) trained on human-engineered features derived from light curves and metadata.

Results. When trained on light curves and metadata, ATAT achieves a macro F1 score of 82.9 ± 0.4 in 20 classes, outperforming the BHRF model trained on 429 features, which achieves a macro F1 score of 79.4 ± 0.1.

Conclusions. The use of transformer multimodal architectures, combining light curves and tabular data, opens new possibilities for classifying alerts from a new generation of large etendue telescopes, such as the Vera C. Rubin Observatory, in real-world brokering scenarios.