Extracting Reened Rules from Knowledge-based Neural Networks Keywords: Theory Reenement Integrated Learning Representational Shift Rule Extraction from Neural Networks

Neural networks, despite their empirically-proven abilities, have been little used for the reenement of existing knowledge because this task requires a three-step process. First, knowledge must be inserted into a neural network. Second, the network must be reened. Third, the reened knowledge must be extracted from the network. We have previously described a method for the rst step of this process. Standard neural learning techniques can accomplish the second step. In this paper, we propose and empirically evaluate a method for the nal, and possibly most diicult, step. Our method eeciently extracts symbolic rules from trained neural networks. The four major results of empirical tests of this method are that the extracted rules: (1) closely reproduce the accuracy of the network from which they are extracted; (2) are superior to the rules produced by methods that directly reene symbolic rules; (3) are superior to those produced by previous techniques for extracting rules from trained neural networks; (4) are \human comprehensible." Thus, this method demonstrates that neural networks can be used to eeectively reene symbolic knowledge. Moreover, the rule-extraction technique developed herein contributes to the understanding of how symbolic and connectionist approaches to artiicial intelligence can be prootably integrated.