# Deep Attention-Based Supernovae Classification of Multi-Band Light-Curves

arXiv: https://arxiv.org/abs/2201.08482 · HTML: https://mlsanity.com/paper/2201.08482/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 57 references that are in the map.
- First posted on arXiv 20 Jan 2022.

## Builds on (most cited first, 10 of 10)

- [RAPID: Early Classification of Explosive Transients using Deep Learning](https://mlsanity.com/paper/1904.00014/) (2019)
- [A Deep Learning Approach for Active Anomaly Detection of Extragalactic Transients](https://mlsanity.com/paper/2103.12102/) (2021)
- [Alert Classification for the ALeRCE Broker System: The Real-time Stamp Classifier](https://mlsanity.com/paper/2008.03309/) (2020)
- [Scalable End-to-end Recurrent Neural Network for Variable star classification](https://mlsanity.com/paper/2002.00994/) (2020)
- [Deep Recurrent Neural Networks for Supernovae Classification](https://mlsanity.com/paper/1606.07442/) (2016)
- [Real-Time Detection of Anomalies in Large-Scale Transient Surveys](https://mlsanity.com/paper/2111.00036/) (2021)
- [Imbalance Learning for Variable Star Classification](https://mlsanity.com/paper/2002.12386/) (2020)
- [Classifying Image Sequences of Astronomical Transients with Deep Neural Networks](https://mlsanity.com/paper/2004.13877/) (2020)
- [Classification of Periodic Variable Stars with Novel Cyclic-Permutation Invariant Neural Networks](https://mlsanity.com/paper/2011.01243/) (2020)
- [The effect of phased recurrent units in the classification of multiple catalogs of astronomical lightcurves](https://mlsanity.com/paper/2106.03736/) (2021)

## Built on by (most cited first, 5 of 5)

- [Transformers for scientific data: a pedagogical review for astronomers](https://mlsanity.com/paper/2310.12069/) (2023)
- [StarEmbed: Benchmarking Time Series Foundation Models on Astronomical Observations of Variable Stars](https://mlsanity.com/paper/2510.06200/) (2025)
- [Identifying Light-curve Signals with a Deep Learning Based Object Detection Algorithm. II. A General Light Curve Classification Framework](https://mlsanity.com/paper/2311.08080/) (2023)
- [Temporal Stamp Classifier: Classifying Short Sequences of Astronomical Alerts](https://mlsanity.com/paper/2405.15073/) (2024)
- [Image-Based Multi-Survey Classification of Light Curves with a Pre-Trained Vision Transformer](https://mlsanity.com/paper/2507.11711/) (2025)

## Abstract (arXiv)

In astronomical surveys, such as the Zwicky Transient Facility, supernovae (SNe) are relatively uncommon objects compared to other classes of variable events. Along with this scarcity, the processing of multi-band light-curves is a challenging task due to the highly irregular cadence, long time gaps, missing-values, few observations, etc. These issues are particularly detrimental to the analysis of transient events: SN-like light-curves. We offer three main contributions: 1) Based on temporal modulation and attention mechanisms, we propose a Deep attention model (TimeModAttn) to classify multi-band light-curves of different SN types, avoiding photometric or hand-crafted feature computations, missing-value assumptions, and explicit imputation/interpolation methods. 2) We propose a model for the synthetic generation of SN multi-band light-curves based on the Supernova Parametric Model, allowing us to increase the number of samples and the diversity of cadence. Thus, the TimeModAttn model is first pre-trained using synthetic light-curves. Then, a fine-tuning process is performed. The TimeModAttn model outperformed other Deep Learning models, based on Recurrent Neural Networks, in two scenarios: late-classification and early-classification. Also, the TimeModAttn model outperformed a Balanced Random Forest (BRF) classifier (trained with real data), increasing the balanced-F1score from ≈.525 to ≈.596. When training the BRF with synthetic data, this model achieved similar performance to the TimeModAttn model proposed while still maintaining extra advantages. 3) We conducted interpretability experiments. High attention scores were obtained for observations earlier than and close to the SN brightness peaks. This also correlated with an early highly variability of the learned temporal modulation.