{"title":"带噪声标签数据的滤波加权校正训练方法","authors":"Yulong Wang, Xiaohui Hu, Zheshu Jia","doi":"10.5220/0010577901770184","DOIUrl":null,"url":null,"abstract":"To solve the problem of low model accuracy under noisy data sets, a filtered weighted correction training method is proposed. This method uses the idea of model fine-tuning to adjust and correct the trained deep neural network model using filtered data, which has high portability. In the data filtering process, the noise label filtering algorithm, which is based on the random threshold in the double interval, reduces the dependence on artificially set parameters, increases the reliability of the random threshold, and improves the filtering accuracy and the recall rate of clean samples. In the calibration process, to deal with sample imbalance, different types of samples are weighted to improve the effectiveness of the model. Experimental results show that the propose method can improve the F1 value of deep neural network model.","PeriodicalId":88612,"journal":{"name":"News. Phi Delta Epsilon","volume":"85 1","pages":"177-184"},"PeriodicalIF":0.0000,"publicationDate":"2021-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Filtered Weighted Correction Training Method for Data with Noise Label\",\"authors\":\"Yulong Wang, Xiaohui Hu, Zheshu Jia\",\"doi\":\"10.5220/0010577901770184\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"To solve the problem of low model accuracy under noisy data sets, a filtered weighted correction training method is proposed. This method uses the idea of model fine-tuning to adjust and correct the trained deep neural network model using filtered data, which has high portability. In the data filtering process, the noise label filtering algorithm, which is based on the random threshold in the double interval, reduces the dependence on artificially set parameters, increases the reliability of the random threshold, and improves the filtering accuracy and the recall rate of clean samples. In the calibration process, to deal with sample imbalance, different types of samples are weighted to improve the effectiveness of the model. Experimental results show that the propose method can improve the F1 value of deep neural network model.\",\"PeriodicalId\":88612,\"journal\":{\"name\":\"News. Phi Delta Epsilon\",\"volume\":\"85 1\",\"pages\":\"177-184\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2021-01-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"News. Phi Delta Epsilon\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.5220/0010577901770184\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"News. Phi Delta Epsilon","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.5220/0010577901770184","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Filtered Weighted Correction Training Method for Data with Noise Label
To solve the problem of low model accuracy under noisy data sets, a filtered weighted correction training method is proposed. This method uses the idea of model fine-tuning to adjust and correct the trained deep neural network model using filtered data, which has high portability. In the data filtering process, the noise label filtering algorithm, which is based on the random threshold in the double interval, reduces the dependence on artificially set parameters, increases the reliability of the random threshold, and improves the filtering accuracy and the recall rate of clean samples. In the calibration process, to deal with sample imbalance, different types of samples are weighted to improve the effectiveness of the model. Experimental results show that the propose method can improve the F1 value of deep neural network model.