{"title":"Extension Of Generalized Delta Rule To Adapt Sigmoid Functions","authors":"A. Sperduti, A. Starita","doi":"10.1109/IEMBS.1991.684507","DOIUrl":null,"url":null,"abstract":"In this paper an extension of the generalized delta rule to adapt sigmoid functions is presented. The proposed learning procedure retains the standard generalized delta rule for weights changes and introduces changes of the sigmoid functions parameters in the same fashion. The adaptation of the nonlinearities so introduced allows to expect an acceleration of the learning and a better adaptation to the input pattern distribution.","PeriodicalId":297811,"journal":{"name":"Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society Volume 13: 1991","volume":"58 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1991-10-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society Volume 13: 1991","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/IEMBS.1991.684507","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
In this paper an extension of the generalized delta rule to adapt sigmoid functions is presented. The proposed learning procedure retains the standard generalized delta rule for weights changes and introduces changes of the sigmoid functions parameters in the same fashion. The adaptation of the nonlinearities so introduced allows to expect an acceleration of the learning and a better adaptation to the input pattern distribution.