| Autor: |
Feldman, Jerome A., Fanty, Mark A., Goddard, Nigel H., Lynne, Kenton J. |
| Předmět: |
|
| Zdroj: |
Communications of the ACM; Feb1988, Vol. 31 Issue 2, p170-187, 18p, 14 Diagrams |
| Abstrakt: |
The article focuses on the use of connectionist/neural networks in automating complex tasks. In computer science, there is a large and growing body of knowledge about parallel computation, and another, largely separate, science of artificial intelligence. The idea of looking directly at massively parallel realizations of intelligent activity promises to be fruitful for the study of both natural and artificial computation. Much attention has been directed toward the biological implications of this interdisciplinary effort,but there are equally important relations with computational theory, hardware and software. From a computational viewpoint, animal brains represent a remarkable machine with properties radically different from those of conventional computers. Connectionist networks consist of simple computational elements that communicate by sending their level of activation via links to other elements. The units have a small number of states and compute simple functions of their inputs. Associated with each link is a weight, indicating the significance of activation arriving over that link. |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
