Common Optimization of Adaptive Preprocessing Units and a Neural Network during the Learning Period. Application in EEG Pattern Recognition

• Published in: Neural networks Vol. 10; no. 6; pp. 1153 - 1163
• Main Authors: Galicki, Miroslaw, Witte, Herbert, Dörschel, Jens, Eiselt, Michael, Griessbach, Gert
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.08.1997; Elsevier Science
• Subjects: Adaptive preprocessing; Applied sciences; Artificial intelligence; Computer science; control theory; systems; Connectionism. Neural networks; EEG patterns; Exact sciences and technology; Feature extraction; Perceptron neural network; Simultaneous training; Feature extraction; Perceptron neural network; EEG patterns; Simultaneous training; Adaptive preprocessing; Learning; Noise reduction; Electroencephalography; Information extraction; Pattern recognition; Neural network; Optimization
• ISSN: 0893-6080; 1879-2782
• DOI: 10.1016/S0893-6080(97)00033-6

In this study, a proposition of simultaneous training of the neural network (multilayer perceptron) and adaptive preprocessing unit is presented. This cooperation enables the network to affect the preprocessing and as a consequence to vary the locations of pattern vectors in a feature space. Thus, during the learning process the network tries to find a good separation of classes of patterns, which results in convergence of the whole learning process. The strategy was developed in order to make efficient EEG monitoring in neonates possible. A comparison of the method presented herein with the known learning strategies for neural networks shows the need for using it as an alternative learning process. The convergence of the whole system is also discussed. © 1997 Elsevier Science Ltd.