Extraction of rules from discrete-time recurrent neural networks
Extraction of rules from discrete-time recurrent neural networks
复制标题
DOI：
10.1016/0893-6080(95)00086-0
复制
发表时间：
1996-01-01
期刊：
影响因子：
7.8
通讯作者：
Giles, CL
中科院分区：
文献类型：
--
作者：
Omlin, CW;Giles, CL
从训练的神经网络中提取符号知识以及在训练之前将（部分）知识直接编码到网络中是重要的问题。它们允许符号和连接主义知识表示之间的信息交换。本文的重点是从递归神经网络中提取的规则的质量。离散时间递归神经网络可以被训练来正确地对正则语言的字符串进行分类。通过在递归状态神经元的输出空间中应用聚类算法，可以以确定性有限状态自动机（DFA）的形式从网络中提取定义学习语法的规则。我们的算法可以从同一个网络中提取出与训练集一致的不同有限状态自动机。我们比较了这些不同的模型和训练的网络的泛化性能，我们引入了一个启发式算法，使我们能够在一致的DFA中选择最接近学习到的正则语法的模型。
The extraction of symbolic knowledge from trained neural networks and the direct encoding of (partial) knowledge into networks prior to training are important issues. They allow the exchange of information between symbolic and connectionist knowledge representations. The focus of the paper is on the quality of the rules that are extracted from recurrent neural networks. Discrete-time recurrent neural networks can be trained to correctly classify strings of a regular language. Rules defining the learned grammar can be extracted from networks in the form of deterministic finite-state automata (DFAs) by applying clustering algorithms in the output space of recurrent state neurons. Our algorithm can extract different finite-state automata that are consistent with a training set from the same network. We compare the generalization performances of these different models and the trained network and we introduce a heuristic that permits us to choose among the consistent DFAs the model which best approximates the learned regular grammar.