Nurul Farhana Mohamad Zamri, N. Md. Tahir, M. S. A. Megat Ali, Nur Dalila Khirul Ashar
{"title":"盗窃检测的深度学习优化算法","authors":"Nurul Farhana Mohamad Zamri, N. Md. Tahir, M. S. A. Megat Ali, Nur Dalila Khirul Ashar","doi":"10.24191/jeesr.v20i1.005","DOIUrl":null,"url":null,"abstract":"34 Abstract— Learning algorithms related to deep learning use bells and whistles, called hyperparameters. Hence, this study conducted numerical analysis, specifically backpropagation gradients and gradient-based optimization for snatch-theft detection. Here, snatch theft images and augmented images were used to perform the experimental study to determine the optimum hyperparameter values. Next, the value of epoch and learning rate was obtained after careful analysis based on each training option. Results achieved showed that epoch value of 20 and learning rate corresponding to 0.0001 was the optimum values. Findings from this study can be used as a practical guide in determining the importance of the most optimum hyperparameters.","PeriodicalId":313365,"journal":{"name":"Journal of Electrical & Electronic Systems Research","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Deep Learning Optimisation Algorithms for Snatch Theft Detection\",\"authors\":\"Nurul Farhana Mohamad Zamri, N. Md. Tahir, M. S. A. Megat Ali, Nur Dalila Khirul Ashar\",\"doi\":\"10.24191/jeesr.v20i1.005\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"34 Abstract— Learning algorithms related to deep learning use bells and whistles, called hyperparameters. Hence, this study conducted numerical analysis, specifically backpropagation gradients and gradient-based optimization for snatch-theft detection. Here, snatch theft images and augmented images were used to perform the experimental study to determine the optimum hyperparameter values. Next, the value of epoch and learning rate was obtained after careful analysis based on each training option. Results achieved showed that epoch value of 20 and learning rate corresponding to 0.0001 was the optimum values. Findings from this study can be used as a practical guide in determining the importance of the most optimum hyperparameters.\",\"PeriodicalId\":313365,\"journal\":{\"name\":\"Journal of Electrical & Electronic Systems Research\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-04-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Electrical & Electronic Systems Research\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.24191/jeesr.v20i1.005\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Electrical & Electronic Systems Research","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.24191/jeesr.v20i1.005","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Deep Learning Optimisation Algorithms for Snatch Theft Detection
34 Abstract— Learning algorithms related to deep learning use bells and whistles, called hyperparameters. Hence, this study conducted numerical analysis, specifically backpropagation gradients and gradient-based optimization for snatch-theft detection. Here, snatch theft images and augmented images were used to perform the experimental study to determine the optimum hyperparameter values. Next, the value of epoch and learning rate was obtained after careful analysis based on each training option. Results achieved showed that epoch value of 20 and learning rate corresponding to 0.0001 was the optimum values. Findings from this study can be used as a practical guide in determining the importance of the most optimum hyperparameters.