法律知识的因果表达和非单调推理模型

Causal reasoning is the most important feature in law consultant systems. This paper analy-ses the structure of law clauses,proposes a representation model for law knowledge in terms of causal relationships and nonmonotonic reasoning models based on it. These models are successfully applied in the implementation of NBU-CALA+ ,a law expert consultant system for case analysis and interpreta-tion.

因果启发的深度域泛化旋转机械故障诊断

针对设备工况变化导致基于深度学习(deep learning,简称DL)的故障诊断性能退化的问题,提出采用因果表示网络(causal representation net,简称CRNet)用于在变工况下实现高性能故障诊断,即域泛化(domain generalization,简称DG)故障诊断。首先,假设DG的结构因果模型,并基于此模型和独立因果机制(independent causal model,简称ICM)原理,得到因果驱动的诊断需求来消除特征间的关联;其次,利用随机傅里叶特征(random Fourier features,简称RFF)将模型提取的特征映射到高维空间,再利用高维空间中的特征构造衡量特征间关联的协方差矩阵,以矩阵非对角值为目标,学习一组权重对样本加权,消除特征间的广义关联;最后,以梯度为引导,屏蔽部分高梯度特征,增强特征包含的诊断信息。锥齿轮传动实验台的实验结果表明,CRNet具备最优的DG性能。