We introduce a novel neural network architecture, referred to as the normalizing neural network (NNN), where the propagated signals take the form of finite probability distributions. Appropriately tuned NNN can be applied as the compound voting measure while classifying new cases on the basis of app
Mapping rule-based systems into neural architecture
β Scribed by Li-Min Fu; Li-Chen Fu
- Publisher
- Elsevier Science
- Year
- 1990
- Tongue
- English
- Weight
- 984 KB
- Volume
- 3
- Category
- Article
- ISSN
- 0950-7051
No coin nor oath required. For personal study only.
π SIMILAR VOLUMES
This is the first of two articles presenting an approach to rule-based expert systems for diagnostic tasks exploiting a purely neural architecture. Here, we outline the methodological options motivating this approach, and describe a forward and backward chaining mechanism on a system of production r
This article describes neural mechanisms for carrying out forms of uncertain reasoning that are frequently applied in diagnostic problem-solving. These include various sorts of causal and probabilistic reasoning, as well as the updating of certainty measures. The control and synchronization function