Space weather and space climate, concerning the conditions and changes in space that impact the Earth and our daily life, is an interdisciplinary science. It involves subjects ranging from the most fundamental, to observing, monitoring and measuring many different systems’ activities and predicting and forecasting their future behaviors.
Recent advancement of artificial intelligence (AI) and machine learning (ML) techniques have led to a variety of new tools for resolving traditional and emerging challenging problems more powerfully from data-driven perspectives. In recent years there has been an increasingly growing application of AI and ML to space weather and space climate problems, including solar activity characterization (e.g. prediction of sunspot cycles), solar wind and geomagnetic field index forecasting, wave mode identification and pattern recognition in the solar atmosphere, and so on.
It is right time to stimulate the dissemination of recent results and findings of applying AI and ML techniques for forecasting space weather and space climate. This Research Topic aims to provide a showcase platform, displaying the recent advancements of AI and ML, as a “learning from data” tool, for dealing with challenging problems in space weather and space climate. One of the primary goals is to facilitate the dissemination of research findings from scientists and researchers who work in the interface and frontier of AI/ML for space physics.
The Research Topic encourages contributions to solving key research questions relating to space weather and space climate using AI and ML techniques, which include, but are not limited to:
1) Forecast of the radiation belts
2) Prediction of geomagnetic indices
3) Application of shallow machine learning and shallow neural networks to space weather and space climate forecasting
e.g., Gaussian process regression (GPR), NARX neural networks ,Radial basis function (RBF) neural networks, Transparent, interpretable and parsimonious models , Signal processing and information theoretic methods , Support vector machines (SVMs) and support vector regression (SVR), Wavelet neural networks
4) Application of deep learning and deep neural networks for space weather and space climate forecasting
e.g., Convolutional neural networks (CNNs), Long short-term memory (LSTM), Recurrent neural networks (RNNs)
5) Interpretable modeling
e.g. System identification, NARX and NARMAX models
6) Explainable machine learning (ML) and artificial intelligence (AI)
e.g. Gradient boosting machines (GBMs), Random forests (RFs)
7) Probability forecasting of space weather and prediction uncertainty analysis
Space weather and space climate, concerning the conditions and changes in space that impact the Earth and our daily life, is an interdisciplinary science. It involves subjects ranging from the most fundamental, to observing, monitoring and measuring many different systems’ activities and predicting and forecasting their future behaviors.
Recent advancement of artificial intelligence (AI) and machine learning (ML) techniques have led to a variety of new tools for resolving traditional and emerging challenging problems more powerfully from data-driven perspectives. In recent years there has been an increasingly growing application of AI and ML to space weather and space climate problems, including solar activity characterization (e.g. prediction of sunspot cycles), solar wind and geomagnetic field index forecasting, wave mode identification and pattern recognition in the solar atmosphere, and so on.
It is right time to stimulate the dissemination of recent results and findings of applying AI and ML techniques for forecasting space weather and space climate. This Research Topic aims to provide a showcase platform, displaying the recent advancements of AI and ML, as a “learning from data” tool, for dealing with challenging problems in space weather and space climate. One of the primary goals is to facilitate the dissemination of research findings from scientists and researchers who work in the interface and frontier of AI/ML for space physics.
The Research Topic encourages contributions to solving key research questions relating to space weather and space climate using AI and ML techniques, which include, but are not limited to:
1) Forecast of the radiation belts
2) Prediction of geomagnetic indices
3) Application of shallow machine learning and shallow neural networks to space weather and space climate forecasting
e.g., Gaussian process regression (GPR), NARX neural networks ,Radial basis function (RBF) neural networks, Transparent, interpretable and parsimonious models , Signal processing and information theoretic methods , Support vector machines (SVMs) and support vector regression (SVR), Wavelet neural networks
4) Application of deep learning and deep neural networks for space weather and space climate forecasting
e.g., Convolutional neural networks (CNNs), Long short-term memory (LSTM), Recurrent neural networks (RNNs)
5) Interpretable modeling
e.g. System identification, NARX and NARMAX models
6) Explainable machine learning (ML) and artificial intelligence (AI)
e.g. Gradient boosting machines (GBMs), Random forests (RFs)
7) Probability forecasting of space weather and prediction uncertainty analysis
