Title :
Using adaptive learning techniques for fast and accurate approximation of physics in numerical models
Author :
Krasnopolsky, Vladimir M. ; Fox-Rabinovitz, Michael ; Chalikov, Dmitry
Author_Institution :
Earth Syst. Sci. Interdisciplinary Center, Maryland Univ., College Park, MD, USA
Abstract :
A new NN application to approximating atmospheric physics processes in numerical climate simulation and weather prediction models is introduced and illustrated.
Keywords :
approximation theory; geophysics computing; learning (artificial intelligence); multilayer perceptrons; numerical analysis; weather forecasting; adaptive learning techniques; atmospheric physics processes; neural networks; numerical climate simulation; numerical models; physics approximation; weather prediction models; Atmospheric modeling; Computational modeling; Geoscience; Machine learning; Neural networks; Numerical models; Physics computing; Predictive models; Supercomputers; Weather forecasting;
Conference_Titel :
Computational Intelligence for Measurement Systems and Applications, 2003. CIMSA '03. 2003 IEEE International Symposium on
Print_ISBN :
0-7803-7783-4
DOI :
10.1109/CIMSA.2003.1227209
