Certain comments on data preparation for neural networks based modelling
The process of data preparation for neural networks based modelling is examined. We are discussing sampling, preprocessing and decimation, finally urguing for orthonormal input preprocessing.
KeywordsNeural Network Data Preparation Linear Dynamical System Dynamic Transformation Neural Network Form
Unable to display preview. Download preview PDF.
- A. Pinkus M. Leshno, V. Lin and S. Schocken. Multilayer feedforward networks with a nonpolynomial activation function can approximate any function. Neural Networks, 13:350–373, 1993.Google Scholar
- L M Grzesiak, B Beliczynski, and B Ufnalski. Input preprocessing in tapped delay neural architecture for induction motor speed estimation. In Proc. EPE Conf. European Power Electronics Conference, Toulouse 2003., EPE, 2003.Google Scholar
- B Beliczynski and L Grzesiak. Induction motor speed estimation: Limitations of certain neural schemes based on time instances of periodic signals. In Proc. IECON’01, pages 653–657. IEEE Service Center, 2001.Google Scholar
- B Beliczynski. On input discretisation processes for tapped delay neural architecture. In Proc. ICANNGA ’03, pages 28–32. Springer, 2003.Google Scholar
- B Beliczynski and K Kubowicz. Appoximately static relationship between model variables. Przeglad Elektrotechniczny, 53(4):385–389, 2004.Google Scholar