“School of Cognitive Sciences”
Back to Papers HomeBack to Papers of School of Cognitive Sciences
| Paper IPM / Cognitive Sciences / 7520 |
|
||||||
| Abstract: | |||||||
|
The Emotional Learning Algorithm, which can be classified in behavioral and reinforcement approaches in artificial intelligence, has been introduced to show the effect of emotions as well known stimuli in the quick and almost satisficing decision making in human. Emotional learning procedure is based on an appropriate response to the emotional signal, which constitutes of multiple criteria without to concern their dimension, and moves toward an acceptable reasoning, providing a satisfactorily good performance in decision making problems. The remarkable properties of emotional learning, low computational complexity and fast training and its simplicity in multi objective problems has made it a powerful methodology in real time control and decision systems, where the gradient based methods and evolutionary algorithms are hard to be used due to their high computational complexity. Recently to emotional approach has been successfully used to obtain multiple criteria in prediction problems of real word phenomena, more specifically space weather forecasting e.g. in predicting solar activity, geomagnetic disturbances and geomagnetic storms.
The solar wind driven magnetosphere is a complex dynamical system with highly nonlinear and chaotic behavior. A large number of studies have been carried out to provide appropriate dynamical models of magnetosphere, and to predict various geomagnetic indices, e.g. Dst storm time index and AE auroral electroject index. But the most popular indicator of geomagnetic disturbances, the KP index, which is used mainly in warning and alert systems for satellites, has not been considered as much. In this study, a recently developed model of emotional learning in human brain is considered to be used in purposeful prediction of KP index and to provide a more reliable K-warning system. The simulated emotional learning procedure inherently emphasizes to learn the features related to high values of Kp related to geomagnetic storms or sub storms. However this learning algorithm is far from an optimal approximation, it is useful to warning and alert systems due to its high rate of correct warning messages in comparison with several other approaches.
Download TeX format |
|||||||
| back to top | |||||||
