Purposeful Prediction of Space Weather Phenomena by Simulated Emotional Learning

Abstract

Bounded rationality and satisfying models, rather than optimization techniques, have shown good performance in decision making. The emotional learning algorithm is an example. It is based on simulating human emotions via reinforcement agents. A new approach towards purposeful prediction problems, derived from a recently developed model of emotional learning in human brain, is introduced in this article. The proposed algorithm inherently emphasizes learning to predict future peaks, and performs remarkably accurate predictions among the important regions, features, or objectives. Space weather forecasting is an excellent example of using this methodology, and in fact was the motivation to introduce purposeful prediction via multiobjective learning algorithm in this research. Three examples of predicting solar activity, geomagnetic activity, and geomagnetic storms show the characteristics of the suggested algorithm and its usefulness to space weather warning and alert systems. The successful application of the proposed model indicates the significance of structural brain modelling beyond traditional neural networks, as well as the importance of biological motivation in choosing a suitable model for any given application.