Title :
Simple heuristic methods for input parameters´ estimation in neural networks
Author :
Tetko, Iaor V. ; Tanchuk, Vsevolod Yu ; Luik, Alexander I.
Author_Institution :
Dept. of Biomed., Inst. of Bioorgan. & Pet. Chem., Kiev, Ukraine
fDate :
27 Jun-2 Jul 1994
Abstract :
We propose simple heuristic methods that can evaluate the relevance of input parameters after completing neural network training. Besides that, these methods allow correct computation of the inputs´ contribution to each problem, when learning multiple tasks simultaneously. The estimations are done on a statistical base and are independent of learning procedures and cost functions. Our simulation on three different tasks shows that these approaches are effective
Keywords :
heuristic programming; learning (artificial intelligence); neural nets; parameter estimation; cost functions; heuristic methods; input parameter estimation; learning; learning procedures; multiple tasks; neural network training; neural networks; simulation; statistical base; Artificial neural networks; Biomedical computing; Chemistry; Computational modeling; Costs; Intelligent networks; Neural networks; Neurons; Parameter estimation; Petroleum;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.374193