# Towards Interpretable Solar Flare Prediction with Attention-based Deep Neural Networks

arXiv: https://arxiv.org/abs/2309.04558 · HTML: https://mlsanity.com/paper/2309.04558/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 42 references that are in the map.
- First posted on arXiv 8 Sep 2023.

## Builds on (most cited first, 12 of 12)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification](https://mlsanity.com/paper/1502.01852/) (2015)
- [One weird trick for parallelizing convolutional neural networks](https://mlsanity.com/paper/1404.5997/) (2014)
- [Learn To Pay Attention](https://mlsanity.com/paper/1804.02391/) (2018)
- [How to Train Your Flare Prediction Model: Revisiting Robust Sampling of Rare Events](https://mlsanity.com/paper/2103.07542/) (2021)
- [Towards Coupling Full-disk and Active Region-based Flare Prediction for Operational Space Weather Forecasting](https://mlsanity.com/paper/2209.07406/) (2022)
- [All-Clear Flare Prediction Using Interval-based Time Series Classifiers](https://mlsanity.com/paper/2105.01202/) (2021)
- [Explaining Full-disk Deep Learning Model for Solar Flare Prediction using Attribution Methods](https://mlsanity.com/paper/2307.15878/) (2023)
- [Explainable Deep Learning-based Solar Flare Prediction with post hoc Attention for Operational Forecasting](https://mlsanity.com/paper/2308.02682/) (2023)
- [Exploring Deep Learning for Full-disk Solar Flare Prediction with Empirical Insights from Guided Grad-CAM Explanations](https://mlsanity.com/paper/2308.15712/) (2023)

## Built on by (most cited first, 7 of 7)

- [Advancing Solar Flare Prediction using Deep Learning with Active Region Patches](https://mlsanity.com/paper/2406.11054/) (2024)
- [Embedding Ordinality to Binary Loss Function for Improving Solar Flare Forecasting](https://mlsanity.com/paper/2408.11768/) (2024)
- [Active Region-based Flare Forecasting with Sliding Window Multivariate Time Series Forest Classifiers](https://mlsanity.com/paper/2402.03474/) (2024)
- [Towards Hybrid Embedded Feature Selection and Classification Approach with Slim-TSF](https://mlsanity.com/paper/2409.04542/) (2024)
- [Large Scale Evaluation of Deep Learning-based Explainable Solar Flare Forecasting Models with Attribution-based Proximity Analysis](https://mlsanity.com/paper/2411.18070/) (2024)
- [Ordinal Encoding as a Regularizer in Binary Loss for Solar Flare Prediction](https://mlsanity.com/paper/2510.04063/) (2025)
- [Enhancing Explainability in Solar Energetic Particle Event Prediction: A Global Feature Mapping Approach](https://mlsanity.com/paper/2511.09475/) (2025)

## Abstract (arXiv)

Solar flare prediction is a central problem in space weather forecasting and recent developments in machine learning and deep learning accelerated the adoption of complex models for data-driven solar flare forecasting. In this work, we developed an attention-based deep learning model as an improvement over the standard convolutional neural network (CNN) pipeline to perform full-disk binary flare predictions for the occurrence of ≥M1.0-class flares within the next 24 hours. For this task, we collected compressed images created from full-disk line-of-sight (LoS) magnetograms. We used data-augmented oversampling to address the class imbalance issue and used true skill statistic (TSS) and Heidke skill score (HSS) as the evaluation metrics. Furthermore, we interpreted our model by overlaying attention maps on input magnetograms and visualized the important regions focused on by the model that led to the eventual decision. The significant findings of this study are: (i) We successfully implemented an attention-based full-disk flare predictor ready for operational forecasting where the candidate model achieves an average TSS=0.54±0.03 and HSS=0.37±0.07. (ii) we demonstrated that our full-disk model can learn conspicuous features corresponding to active regions from full-disk magnetogram images, and (iii) our experimental evaluation suggests that our model can predict near-limb flares with adept skill and the predictions are based on relevant active regions (ARs) or AR characteristics from full-disk magnetograms.