人工智能培训

搜索

机器学习论文:SECUREBOOST:无损联邦学习框架(SecureBoost: A Lossless Federated Learning Framework

[复制链接]
kjdshfjsfgsfg 发表于 2019-1-28 11:56:42 | 显示全部楼层 |阅读模式
kjdshfjsfgsfg 2019-1-28 11:56:42 58 0 显示全部楼层
机器学习论文:SECUREBOOST:无损联邦学习框架(SecureBoost: A Lossless Federated Learning Framework)保护用户隐私是机器学习中的一个重要问题,正如2018年5月在欧盟(EU)推出通用数据保护法规(GDPR)所证明的那样.GDPR旨在让用户更好地控制他们的个人数据,这促使我们在不违反用户隐私的情况下通过数据共享来探索机器学习框架。为了实现这一目标,在本文中,我们提出了一种新的无损隐私保护树增强系统,称为SecureBoost,用于教学设置。该联合学习系统允许学习过程通过具有部分公共用户样本但不同特征集的多方联合进行,其对应于垂直分区的虚拟数据集。 SecureBoost的一个优点是它提供与非隐私保护方法相同的精度,同时不会泄露每个私有数据提供者的信息。我们理论上证明SecureBoost框架与将数据集中到一个地方的其他非联合梯度树提升算法一样准确。此外,除了安全性证明之外,我们还讨论了使协议完全安全所需的条件。
The protection of user privacy is an important concern in machine learning,as evidenced by the rolling out of the General Data Protection Regulation(GDPR) in the European Union (EU) in May 2018. The GDPR is designed to giveusers more control over their personaldata, which motivates us to exploremachine learning frameworks with data sharing without violating user privacy.To meet this goal, in this paper, we propose a novel losslessprivacy-preserving tree-boosting system known as SecureBoost in the setting offederated learning.This federated-learning system allows a learning process tobe jointly conducted over multiple parties with partially common user samplesbut different feature sets, which corresponds to a vertically partitionedvirtual data set.An advantage of SecureBoost is that it provides the samelevel of accuracy as the non-privacy-preserving approach while at the sametime, reveal no information of each private data provider.We theoreticallyprove that the SecureBoost framework is as accurate as other non-federatedgradient tree-boosting algorithms that bring the data into one place.Inaddition, along with a proof of security, we discuss what would be required tomake the protocols completely secure.机器学习论文:SECUREBOOST:无损联邦学习框架(SecureBoost: A Lossless Federated Learning Framework) k6tpyT6UIKw7p6n2.jpg
URL地址:https://arxiv.org/abs/1901.08755     ----pdf下载地址:https://arxiv.org/pdf/1901.08755    ----机器学习论文:SECUREBOOST:无损联邦学习框架(SecureBoost: A Lossless Federated Learning Framework)
回复

使用道具 举报

您需要登录后才可以回帖 登录 | 立即注册

本版积分规则 返回列表 发新帖

kjdshfjsfgsfg当前离线
新手上路

查看:58 | 回复:0

快速回复 返回顶部 返回列表