Is designing a neural network application an art or a science?

Authors:
Roman Erenshteyn;Richard Foulds;Scott Galuska
Affiliations:
-;-;-
Venue:
ACM SIGCHI Bulletin
Year:
1994

Citing 3
Cited 0

Neural network models for pattern recognition and associative memory

Neural Networks
Learning internal representations by error propagation

Parallel distributed processing: explorations in the microstructure of cognition, vol. 1
Artificial Neural Networks: Paradigms, Applications, and Hardware Implementations

Artificial Neural Networks: Paradigms, Applications, and Hardware Implementations

Quantified Score

Hi-index	0.00

Visualization

Abstract

Both the danger and the attraction of pattern recognition problems lie in their simplicity. When investigators first approach such a problem, they may, with an easiness which borders on carelessness, utilize the methods most familiar to them; and... they sometimes receive satisfactory results. More often, they are haunted by troubles. Only a good understanding of the problem will permit them to solve it. The quality of problem solving depends on the investigator's abilities, experience, knowledge and very often on... fertile imagination. In this paper we will discuss the problems connected with practical neural network application design.