{"title":"卷积神经网络简化的冗余特征检测与去除","authors":"Shih-Chang Hsia, Yuedong Yang","doi":"10.1109/ICCCI51764.2021.9486779","DOIUrl":null,"url":null,"abstract":"Since the rapid development of GPUs, the AI model of the convolutional neural network (CNN) has also made great progress. Researchers have gradually developed the model in a deeper and wider direction, hoping to have better accuracy. Although this is indeed effective, it also causes the model has too many parameters, and it takes a lot of time to calculate. In such a complex model, some operations are no effect on the output results. In this paper, we use several methods to remove the less important operations from the CNN model. This algorithm can reduce the amount of parameters and calculations while maintaining accuracy.","PeriodicalId":180004,"journal":{"name":"2021 3rd International Conference on Computer Communication and the Internet (ICCCI)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2021-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Redundancy Features Detection and Removal for Simplification of Convolutional Neural Networks\",\"authors\":\"Shih-Chang Hsia, Yuedong Yang\",\"doi\":\"10.1109/ICCCI51764.2021.9486779\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Since the rapid development of GPUs, the AI model of the convolutional neural network (CNN) has also made great progress. Researchers have gradually developed the model in a deeper and wider direction, hoping to have better accuracy. Although this is indeed effective, it also causes the model has too many parameters, and it takes a lot of time to calculate. In such a complex model, some operations are no effect on the output results. In this paper, we use several methods to remove the less important operations from the CNN model. This algorithm can reduce the amount of parameters and calculations while maintaining accuracy.\",\"PeriodicalId\":180004,\"journal\":{\"name\":\"2021 3rd International Conference on Computer Communication and the Internet (ICCCI)\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-06-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2021 3rd International Conference on Computer Communication and the Internet (ICCCI)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/ICCCI51764.2021.9486779\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 3rd International Conference on Computer Communication and the Internet (ICCCI)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/ICCCI51764.2021.9486779","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Redundancy Features Detection and Removal for Simplification of Convolutional Neural Networks
Since the rapid development of GPUs, the AI model of the convolutional neural network (CNN) has also made great progress. Researchers have gradually developed the model in a deeper and wider direction, hoping to have better accuracy. Although this is indeed effective, it also causes the model has too many parameters, and it takes a lot of time to calculate. In such a complex model, some operations are no effect on the output results. In this paper, we use several methods to remove the less important operations from the CNN model. This algorithm can reduce the amount of parameters and calculations while maintaining accuracy.