神经网络设计的构造和剪枝方法

VII Brazilian Symposium on Neural Networks, 2002. SBRN 2002. Proceedings. Pub Date : 2002-11-11 DOI:10.1109/SBRN.2002.1181434

M. Costa, A. Braga, B. R. Menezes

{"title":"神经网络设计的构造和剪枝方法","authors":"M. Costa, A. Braga, B. R. Menezes","doi":"10.1109/SBRN.2002.1181434","DOIUrl":null,"url":null,"abstract":"This paper presents methods to improve generalization of multilayer perceptron (MLP) by pruning the original topology without loss in performance. Topology information and validation sets are used. The results show that these techniques are able to choose a minimum network topology and to simplify trained networks.","PeriodicalId":157186,"journal":{"name":"VII Brazilian Symposium on Neural Networks, 2002. SBRN 2002. Proceedings.","volume":"13 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2002-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"10","resultStr":"{\"title\":\"Constructive and pruning methods for neural network design\",\"authors\":\"M. Costa, A. Braga, B. R. Menezes\",\"doi\":\"10.1109/SBRN.2002.1181434\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"This paper presents methods to improve generalization of multilayer perceptron (MLP) by pruning the original topology without loss in performance. Topology information and validation sets are used. The results show that these techniques are able to choose a minimum network topology and to simplify trained networks.\",\"PeriodicalId\":157186,\"journal\":{\"name\":\"VII Brazilian Symposium on Neural Networks, 2002. SBRN 2002. Proceedings.\",\"volume\":\"13 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2002-11-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"10\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"VII Brazilian Symposium on Neural Networks, 2002. SBRN 2002. Proceedings.\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/SBRN.2002.1181434\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"VII Brazilian Symposium on Neural Networks, 2002. SBRN 2002. Proceedings.","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SBRN.2002.1181434","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 10

摘要

本文提出了在不损失性能的前提下，通过对多层感知器(MLP)的原始拓扑进行剪枝来提高其泛化性能的方法。使用拓扑信息和验证集。结果表明，这些技术能够选择最小的网络拓扑并简化训练后的网络。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Constructive and pruning methods for neural network design

This paper presents methods to improve generalization of multilayer perceptron (MLP) by pruning the original topology without loss in performance. Topology information and validation sets are used. The results show that these techniques are able to choose a minimum network topology and to simplify trained networks.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

VII Brazilian Symposium on Neural Networks, 2002. SBRN 2002. Proceedings.

自引率

0.00%

发文量