一句话：提出DENAS，从训练好的DNN中提取知识自动生成规则，以弥合DNN与规则系统间的差距。
深度神经网络（DNN）已被广泛用于从海量数据中自动学习模式，但在安全或安保关键场景中，许多应用仍基于领域专家手工编写并验证的规则进行决策。为弥合DNN与规则系统之间的差距，本文提出DENAS，通过从训练良好的DNN中提取知识来自动化规则生成。现有同类技术要么依赖特定DNN输入实例，要么采用本质上不稳定的输入空间随机采样，因此仅将探索限制在DNN的局部决策空间，或无法收敛到一致的规则集合，导致生成规则缺乏代表性与稳定性。摘要未报告具体方法细节与实验评估结果。
Deep neural networks (DNNs) have been widely applied in the software development process to automatically learn patterns from massive data. However, many applications still make decisions based on rules that are manually crafted and verified by domain experts due to safety or security concerns. In this paper, we aim to close the gap between DNNs and rule-based systems by automating the rule generation process via extracting knowledge from well-trained DNNs. Existing techniques with similar purposes either rely on specific DNNs input instances or use inherently unstable random sampling of the input space. Therefore, these approaches either limit the exploration area to a local decision-space of the DNNs or fail to converge to a consistent set of rules. The resulting rules thus lack representativeness and stability.
DOI 原文 ·
@inproceedings{paperbot5864,
title = {DENAS: automated rule generation by knowledge extraction from neural networks},
author = {Simin Chen and Soroush Bateni and Sampath Grandhi and Xiaodi Li and Cong Liu and Wei Yang},
booktitle = {FSE},
year = {2020},
doi = {10.1145/3368089.3409733}
}