Title :
Gaussian network variants: a preliminary study
Author :
Koffman, Stephen J. ; Mechl, P.H.
Author_Institution :
Sch. of Mech. Eng., Purdue Univ., West Lafayette, IN, USA
Abstract :
Five variants of a Gaussian network are compared with respect to learning capabilities on three different data sets. The variations allow different degrees of flexibility in approximating a function. Greater flexibility is incurred at the cost of increased processing time. Simulation results indicate a point of diminishing returns with increasing flexibility
Keywords :
function approximation; learning (artificial intelligence); neural nets; Gaussian network; data sets; function approximation; learning capabilities; processing time; Costs; Equations; Intelligent networks; Mechanical engineering; System identification;
Conference_Titel :
Neural Networks, 1993., IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0999-5
DOI :
10.1109/ICNN.1993.298612
