首页 | 本学科首页   官方微博 | 高级检索  
     

基于差分进化生物地理学优化的多层感知器训练方法
引用本文:王娟,吴宪祥,曹艳玲. 基于差分进化生物地理学优化的多层感知器训练方法[J]. 计算机应用研究, 2017, 34(3)
作者姓名:王娟  吴宪祥  曹艳玲
作者单位:西安翻译学院工程技术学院,西安电子科技大学空间科学与技术学院ICIE研究所,西安电子科技大学空间科学与技术学院ICIE研究所
基金项目:国家自然科学基金资助项目 (61105066);中央高校基本科研业务费专项资金资助项目(JB141305)
摘    要:针对生物地理学优化训练多层感知器存在的早熟收敛以及初始化灵敏等问题,提出一种基于差分进化生物地理学优化的多层感知器训练方法。将生物地理学优化(Biogeography-based Optimization,BBO)与差分进化(Differential Evolution,DE)算法相结合,形成改进的混合DE_BBO算法;采用改进的DE_BBO来训练多层感知器(Multi-Layer Perceptron,MLP),并应用于虹膜、乳腺癌、输血、钞票验证等4类数据分类。与BBO、PSO、GA、ACO、ES、PBIL等6种主流启发式算法的实验结果进行比较表明,DE_BBO_MLP算法在分类精度和收敛速度等方面优于已有方法。

关 键 词:生物地理学优化;差分进化;多层感知器;数据分类
收稿时间:2015-09-29
修稿时间:2017-01-16

Multi-Layer Perceptron Using Hybrid Differential Evolution and Biogeography-based Optimization
WANG Juan,WU Xian-xiang and CAO Yan-ling. Multi-Layer Perceptron Using Hybrid Differential Evolution and Biogeography-based Optimization[J]. Application Research of Computers, 2017, 34(3)
Authors:WANG Juan  WU Xian-xiang  CAO Yan-ling
Abstract:The problems of premature convergence and initialization-sensitive are often experiencing when we train the Multi-Layer Perceptron using the biogeography-based optimization. This paper proposed a novel Multi-Layer Perceptron training method using hybrid differential evolution and biogeography-based optimization. This paper introduces the differential evolution to the biogeography-based optimization to construct the hybrid DE_BBO algorithm and then uses the hybrid DE_BBO algorithm for training MLPs. In order to investigate the efficiencies of DE_BBO in training MLPs, this paper employed four classification datasets, including the Iris dataset, the Breast cancer dataset, the Blood Transfusion datasets and the Banknote authentication dataset. Comparing with six well-known heuristic algorithms, including BBO, PSO, GA, ACO, ES, and PBIL in a statistically significant way, the experimental results show that training MLPs using hybrid DE_BBO is significantly better than the current heuristic learning algorithms in terms of convergence speed and convergence accuracy.
Keywords:
点击此处可从《计算机应用研究》浏览原始摘要信息
点击此处可从《计算机应用研究》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