首页 | 本学科首页   官方微博 | 高级检索  
     

基于规则推理网络的分类模型
引用本文:黄德根,张云霞,林红梅,邹丽,刘壮.基于规则推理网络的分类模型[J].软件学报,2020,31(4):1063-1078.
作者姓名:黄德根  张云霞  林红梅  邹丽  刘壮
作者单位:大连理工大学计算机科学与技术学院,辽宁大连 116024;辽宁师范大学计算机与信息技术学院,辽宁大连 116081
基金项目:国家自然科学基金(61772250,61672127)
摘    要:为了缓解神经网络的“黑盒子”机制引起的算法可解释性低的问题,基于使用证据推理算法的置信规则库推理方法(以下简称RIMER)提出了一个规则推理网络模型.该模型通过RIMER中的置信规则和推理机制提高网络的可解释性.首先证明了基于证据推理的推理函数是可偏导的,保证了算法的可行性;然后,给出了规则推理网络的网络框架和学习算法,利用RIMER中的推理过程作为规则推理网络的前馈过程,以保证网络的可解释性;使用梯度下降法调整规则库中的参数以建立更合理的置信规则库,为了降低学习复杂度,提出了“伪梯度”的概念;最后,通过分类对比实验,分析了所提算法在精确度和可解释性上的优势.实验结果表明,当训练数据集规模较小时,规则推理网络的表现良好,当训练数据规模扩大时,规则推理网络也能达到令人满意的结果.

关 键 词:规则推理  RIMER  可解释性网络  机器学习  不确定性分类
收稿时间:2019/3/10 0:00:00
修稿时间:2019/7/11 0:00:00

Rule Inference Network Model for Classification
HUANG De-Gen,ZHANG Yun-Xi,LIN Hong-Mei,ZOU Li,LIU Zhuang.Rule Inference Network Model for Classification[J].Journal of Software,2020,31(4):1063-1078.
Authors:HUANG De-Gen  ZHANG Yun-Xi  LIN Hong-Mei  ZOU Li  LIU Zhuang
Affiliation:School of Computer Science and Technology, Dalian University of Technology, Dalian 116024, China,School of Computer Science and Technology, Dalian University of Technology, Dalian 116024, China,School of Computer and Information Technology, Liaoning Normal University, Dalian 116081, China,School of Computer and Information Technology, Liaoning Normal University, Dalian 116081, China and School of Computer Science and Technology, Dalian University of Technology, Dalian 116024, China
Abstract:The black-box working mechanism of artificial neutral network brings the confusion of interpretability. We propose a rule inference network based on rule-base inference methodology using the evidential reasoning approach (RIMER). It is interpretable by the rules and the inference engine in RIMER. In the present work, the partial derivatives of inference functions are proved as the theoretical fundamental of the proposed model. The framework and the learning algorithm of rule inference network for classification are presented. The feedforward of rule inference network using the inference process in RIMER, contributes for the interpretability. Meanwhile, parameters in belief rule base such as attribute weights, rule weights and belief degree of consequents are trained by gradient descent as in neural network for belief rule base establishment. Moreover, we simplify the gradient by proposing the "pseudo gradient" to reduce the learning complex during the training process. The experiment results indicate the advantages of proposed rule inference network on both interpretability and learning capability. It shows that the rule inference network performs well when the scale of the training dataset is small, and when the training data scale increases, it also achieves comforting results.
Keywords:Rule inference network  RIMER  Interpretable network  Machine learning  Uncertainty classification
本文献已被 维普 万方数据 等数据库收录!
点击此处可从《软件学报》浏览原始摘要信息
点击此处可从《软件学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号