支持规则推理的神经网络模型

本文研究了支持规则推理的神经网络模型，表明通常执行的推理与符号系统在方法上确实相似，只是它们对常识推理提供了更多的方法。CONSYDERR是一种支持常识推理的连接结构，其目的是给出常识推理的一种模型，并纠正传统规则系统中的脆弱性问题。本项工作表明，推理的连接模型不仅实现了符号推理，而且是一种更好的常识推理的计算模型。

Neural Network Models for Rule-Based Reasoning

Neural network models of rule-based reasoning are investigated and it is shown that while such models usually carry out reasoning in exactiy the same way as symbolic systems, they have more to offer in im of commonsense reasoning. CONSYDRR,is proposed to account fOo commonsense reasoning patterns and to remedy the brittleness problem in traditional rule-based systems. This work sbows that connection models of reasoning are not just implementatiou of their symbolic counterparts, but computational models of commonsense reasoning.