针对训练数据错误构建弹性集成

2022 52nd Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W) Pub Date : 2022-06-01 DOI:10.1109/dsn-w54100.2022.00020

Abraham Chan, A. Gujarati, K. Pattabiraman, S. Gopalakrishnan

{"title":"针对训练数据错误构建弹性集成","authors":"Abraham Chan, A. Gujarati, K. Pattabiraman, S. Gopalakrishnan","doi":"10.1109/dsn-w54100.2022.00020","DOIUrl":null,"url":null,"abstract":"In this talk, we describe our approach to construct resilient ML ensembles against training data faults [1]. First, we demonstrate how ensembles tolerate faulty training data. Then, we show how we could use analytical modelling to help ML practitioners build resilient ensembles without the need for resource intensive fault injection experiments.","PeriodicalId":349937,"journal":{"name":"2022 52nd Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Towards Building Resilient Ensembles against Training Data Faults\",\"authors\":\"Abraham Chan, A. Gujarati, K. Pattabiraman, S. Gopalakrishnan\",\"doi\":\"10.1109/dsn-w54100.2022.00020\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In this talk, we describe our approach to construct resilient ML ensembles against training data faults [1]. First, we demonstrate how ensembles tolerate faulty training data. Then, we show how we could use analytical modelling to help ML practitioners build resilient ensembles without the need for resource intensive fault injection experiments.\",\"PeriodicalId\":349937,\"journal\":{\"name\":\"2022 52nd Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)\",\"volume\":\"1 1\",\"pages\":\"0\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-06-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"2022 52nd Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.1109/dsn-w54100.2022.00020\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"2022 52nd Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/dsn-w54100.2022.00020","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

摘要

在这次演讲中，我们描述了我们构建弹性ML集成来对抗训练数据错误的方法[1]。首先，我们演示了集成系统如何容忍错误的训练数据。然后，我们展示了如何使用分析建模来帮助ML从业者在不需要资源密集型故障注入实验的情况下构建弹性集成。

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

Towards Building Resilient Ensembles against Training Data Faults

In this talk, we describe our approach to construct resilient ML ensembles against training data faults [1]. First, we demonstrate how ensembles tolerate faulty training data. Then, we show how we could use analytical modelling to help ML practitioners build resilient ensembles without the need for resource intensive fault injection experiments.

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

2022 52nd Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W)

自引率

0.00%

发文量