Title :
Physics without laws-making exact predictions with data based methods
Author :
Kindermann, Lars ; Protzel, Peter
Author_Institution :
Lab. for Math. Neurosci., RIKEN Brain Sci. Inst., Japan
fDate :
6/24/1905 12:00:00 AM
Abstract :
The mathematical method of fractional or continuous iteration can be used to model a dynamical system exactly from limited experimental data. However, mathematics is complicated and exact solutions-even if proven to exist-can rarely be found analytically. We have shown previously that neural networks can be utilized to numerically compute fractional iterates of mathematical functions. In this paper we demonstrate the application of this method to the fundamental experiment of physics: the free fall
Keywords :
iterative methods; multilayer perceptrons; physics; continuous iteration; data based methods; dynamical system; exact predictions; fractional iteration; free fall experiment; Brain modeling; Data mining; Equations; Mathematical model; Mathematics; Neuroscience; Physics; Predictive models; Training data; World Wide Web;
Conference_Titel :
Neural Networks, 2002. IJCNN '02. Proceedings of the 2002 International Joint Conference on
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-7278-6
DOI :
10.1109/IJCNN.2002.1007769
