We define the stable semantics for general hypothetical logic programs. We consider resolving a hypothetical goal (G ∶ R) in a context P as consisting of two steps: (i) Updating the context by inserting the clauses i...
详细信息
We give an account of various semantics for hierarchical logic programs and discuss their properties in terms of compositionality and full abstraction when inheritance is assumed as the underlying composition mechanis...
详细信息
In this paper we propose a logic-based language, CLP (AD) which is an instance of the Constraint logicprogramming schema and is a convenient semantic framework to be used for deductive database language with updates....
详细信息
This paper attempts a direct semantic formalization of first-order relational-functional languages (the characteristic RELFUN subset) in terms of a generalized model concept. Function-defining conditional equations (o...
详细信息
We present a new model for concurrent programming, which we call AbstrAct. A system is specified through collections of rules defining all possible state transformations that may occur. The activity of a system is dri...
详细信息
Neural networks have proven very useful in the field of pattern classification by mapping input patterns into one of several categories. The most popular methods of training neural networks use gradient descent techni...
详细信息
ISBN:
(纸本)1565550072
Neural networks have proven very useful in the field of pattern classification by mapping input patterns into one of several categories. The most popular methods of training neural networks use gradient descent techniques to minimize mean square error measured against a training set. The composition of this training set, therefore, will determine the performance of the resulting trained network. Deficiencies in the training set, such as unequal class representation or ineffective training vectors will diminish the network's ultimate performance. Additionally, the characteristics of the training set will implicitly set network performance goals that may or may not match the desired goals of the network designer. This paper investigates the utilization of a fuzzy system to overcome these training set inadequacies by incorporating performance goals into the training strategy. Fuzzy control is used to modify the learning rate parameter during backpropagation training. The method is tested using artificial training data and real-world hyperspectral imagery.
Modal logic can be regarded as a unifying framework in which different logical languages with blocks and modules can be expressed. In particular, a modal characterization can be given to different languages which are ...
详细信息
The purpose of this work is to define a theorem prover that retains the procedural aspects of logic programing. The proof system we propose (SLWV-resolution, for Selected Linear Without contrapositive clause Variants)...
详细信息
A way of introducing simple (finite) set designations and operations as firstclass objects of an (unrestricted) logicprogramming language is discussed from both the declarative and the operational semantics viewpoint...
详细信息
In this paper we present a narrower for conditional equational theories whose clauses allow disequations in their bodies (normal theories). Our approach deals with disequations in a constructive manner and thus allows...
详细信息
暂无评论