Computational & Technology Resources
an online resource for computational,
engineering & technology publications |
|
Civil-Comp Proceedings
ISSN 1759-3433 CCP: 16
NEURAL NETWORKS & COMBINATORIAL OPTIMIZATION IN CIVIL & STRUCTURAL ENGINEERING Edited by: B.H.V. Topping and A.I. Khan
Paper II.1
Common Misconceptions About Neural Networks as Approximators W.C. Carpenter* and J-F. Barthelemy+
*Department of Civil Engineering and Mechanics, University of South Florida, Tampa, Florida, United States of America
W.C. Carpenter, J-F. Barthelemy, "Common Misconceptions About Neural Networks as Approximators", in B.H.V. Topping, A.I. Khan, (Editors), "Neural Networks & Combinatorial Optimization in Civil & Structural Engineering", Civil-Comp Press, Edinburgh, UK, pp 11-18, 1993. doi:10.4203/ccp.16.2.1
Abstract
A current trend in scientific and engineering computing is to use neural network approximations instead of polynomial
approximations or other types of approximations involving mathematical functions. A number of misconceptions have arisen
concerning neural networks as approximations. The goal of this paper is to eliminate these misconceptions. In so doing, the
paper examines the computational efficiency of neural network approximations compared to polynomial approximations,
examines the effect of using under-determined neural network approximations, examines the effect of design selection on the
quality of neural network approximations, examines the effect of over-training neural networks, and examines the computing
time required to train neural networks compared to the time to develop polynomial approximations.
purchase the full-text of this paper (price £20)
go to the previous paper |
|