A Machine Learning Approach to Surveying Solar Energetic Particle Time Profiles

We have pursued machine learning (ML) techniques to analyze solar energetic particle (SEP) data as supplements to traditional algorithmic analysis. Machine learning approaches have the potential to execute large surveys of SEP event data much faster than traditional algorithmic approaches and to identify features in large data sets that are unseen by traditional analysis without a priori knowledge of their existence. In this presentation, we apply classifiers to SEP time profiles to distinguish between SEP events with exponential decays (suggesting convective transport and adiabatic deceleration), power law decays (suggesting diffusive propagation), and irregular profile shapes, and we apply regressors to extract time decay constants from fittable decays. These time decay constants can also be used for broad surveys of inferred ionic charge states in SEP events using the technique of Sollitt et al. (2008). We have trained ML models with ACE/SIS and STEREO/LET data, with target data / labels obtained by algorithmic fits to identifiable exponential decays, and preliminary tests have yielded exponential decay classification with >90% accuracy to date. These time profiles include elements and energies available in ACE/SIS spacecraft data since the ACE spacecraft launched in 1997 (e.g. He, CNO, Ne, Mg, Si, and Fe, at energies up to 168 MeV/nuc for Fe). We have also completed initial regression analysis with ACE/SIS and STEREO/LET data, to obtain exponential decays using trained ML regressors.