<?xml version="1.0" encoding="utf-8" ?><rss version="2.0"><channel><title><![CDATA[WINCJS的博客]]></title><description><![CDATA[]]></description><link>https://blog.youkuaiyun.com/WINCJS</link><language>zh-cn</language><generator>https://blog.youkuaiyun.com/</generator><copyright><![CDATA[Copyright &copy; WINCJS]]></copyright><item><title><![CDATA[论文学习笔记 Titanium: A Metadata-Hiding File-Sharing System with Malicious Security]]></title><link>https://blog.youkuaiyun.com/WINCJS/article/details/127851242</link><guid>https://blog.youkuaiyun.com/WINCJS/article/details/127851242</guid><author>WINCJS</author><pubDate>Mon, 14 Nov 2022 17:25:29 +0800</pubDate><description><![CDATA[安全顶会 NDSS 2022论文，一种具有恶意安全的元数据隐藏文件分享系统，使用技术包括：安全多方计算MPC、Schwartz-Zippel Lemma、秘密分析secret sharing、Circuit ORAM等
]]></description><category></category></item><item><title><![CDATA[论文学习笔记 SIMC: ML Inference Secure Against Malicious Clients at Semi-Honest Cost]]></title><link>https://blog.youkuaiyun.com/WINCJS/article/details/124146351</link><guid>https://blog.youkuaiyun.com/WINCJS/article/details/124146351</guid><author>WINCJS</author><pubDate>Wed, 13 Apr 2022 16:18:35 +0800</pubDate><description><![CDATA[安全顶级会议USENIX2022年，恶意半诚实模型下安全神经网络推理协议。混淆电路、同态加密、认证分享、不经意传输。]]></description><category></category></item><item><title><![CDATA[论文学习笔记 MUSE: Secure Inference Resilient to Malicious Clients]]></title><link>https://blog.youkuaiyun.com/WINCJS/article/details/124039456</link><guid>https://blog.youkuaiyun.com/WINCJS/article/details/124039456</guid><author>WINCJS</author><pubDate>Fri, 08 Apr 2022 15:49:04 +0800</pubDate><description><![CDATA[该论文是2021年信息安全顶会USENIX上录用的文章之一。该文章首先针对现有半诚实安全神经网络推理进行了模型提取攻击，然后提出了在恶意用户模型下安全的神经网络推理协议MUSE。]]></description><category></category></item><item><title><![CDATA[论文学习笔记 PrivateDrop: Practical Privacy-Preserving Authentication for Apple AirDrop]]></title><link>https://blog.youkuaiyun.com/WINCJS/article/details/120623152</link><guid>https://blog.youkuaiyun.com/WINCJS/article/details/120623152</guid><author>WINCJS</author><pubDate>Wed, 06 Oct 2021 13:17:53 +0800</pubDate><description><![CDATA[论文学习笔记 PrivateDrop: Practical Privacy-Preserving  Authentication for Apple AirDrop背景一、AirDrop协议二、隐私问题发送方S联系人识别号泄漏接收方R联系人识别号泄漏三、PrivateDrop：基于PSI的双向认证协议设计选项和最终设计PSI协议选择基于PSI的PrivateDrop协议总结

背景
 Apple AirDrop隔空投送的双向认证过程中存在隐私问题，即敌手会获得发送方或者接收方的联系人识别号(Contact.]]></description><category></category></item><item><title><![CDATA[论文学习笔记 POSEIDON: Privacy-Preserving Federated Neural Network Learning]]></title><link>https://blog.youkuaiyun.com/WINCJS/article/details/118436200</link><guid>https://blog.youkuaiyun.com/WINCJS/article/details/118436200</guid><author>WINCJS</author><pubDate>Sat, 03 Jul 2021 11:50:42 +0800</pubDate><description><![CDATA[论文学习笔记 POSEIDON: Privacy-Preserving Federated Neural Network Learning
NDSS 2021录用文章

目录论文学习笔记 POSEIDON: Privacy-Preserving Federated Neural Network Learning一、机器学习1. 机器学习(ML)中的挑战2. 隐私保护机器学习(PPML)二、POSEIDON方案1. 系统和威胁模型2. 方案总览多方同态加密(MHE)联邦学习主要挑战和解决方法3. 方案CKKS]]></description><category></category></item><item><title><![CDATA[论文学习笔记 Diogenes: Lightweight Scalable  RSA Modulus Generation with a  Dishonest Majority]]></title><link>https://blog.youkuaiyun.com/WINCJS/article/details/115234357</link><guid>https://blog.youkuaiyun.com/WINCJS/article/details/115234357</guid><author>WINCJS</author><pubDate>Fri, 26 Mar 2021 17:13:34 +0800</pubDate><description><![CDATA[论文学习笔记 Diogenes: Lightweight Scalable  RSA Modulus Generation with a  Dishonest MajorityIntroductionSchemeGoalStep 1: Scalable  Passive ProtocolBoneh-Franklin Framework[BF97]Overview of techniquesApproachThreshold Homomorphic EncryptionGeneration process[B]]></description><category></category></item><item><title><![CDATA[论文学习笔记 How not to prove your election outcome]]></title><link>https://blog.youkuaiyun.com/WINCJS/article/details/109721851</link><guid>https://blog.youkuaiyun.com/WINCJS/article/details/109721851</guid><author>WINCJS</author><pubDate>Mon, 16 Nov 2020 15:40:15 +0800</pubDate><description><![CDATA[论文学习笔记 How not to prove your election outcome一、投票系统的错误总结二、投票系统介绍三、密钥说明四、个人可验证性阶段协议五、FIAT-SHAMIR转换的陷阱：提供虚假的选票有效性证明六、客户端证明不充足：当选择码与选票不匹配时，生成选票有效性的证明七、单独可验证性：为操作的投票返回正确的选择码（服务器勾结的情况下）八、使用BAYER-GROTH中的陷门承诺和伪造shuffle证明8.1 BAYER-GROTH承诺8.2 使用BAYER-GROTH中的陷门承诺和伪造]]></description><category></category></item><item><title><![CDATA[论文学习笔记 DELPHI: A Cryptographic Inference Service for Neural Networks]]></title><link>https://blog.youkuaiyun.com/WINCJS/article/details/107669375</link><guid>https://blog.youkuaiyun.com/WINCJS/article/details/107669375</guid><author>WINCJS</author><pubDate>Wed, 29 Jul 2020 16:44:52 +0800</pubDate><description><![CDATA[论文学习笔记 DELPHI: A Cryptographic Inference Service for Neural Networks背景介绍出发点系统概括预处理阶段在线阶段DELPHI 规划器
背景
许多公司为用户提供神经网络预测服务，比如说，家庭监控系统，使用专有的神经网络对客户家中的监控视频中的对象进行分类，例如停在用户家附近的汽车，或访客的人脸。但是，当前的预测系统损害了用户和公司中其中一方的隐私：要么用户必须将敏感输入发送给服务提供商进行分类，损害了用户的个人隐私。要么服务提供商必须将其专有的神]]></description><category></category></item><item><title><![CDATA[论文学习笔记 MemGuard: Defending against Black-Box Membership Inference Attacks via Adversarial Examples]]></title><link>https://blog.youkuaiyun.com/WINCJS/article/details/105763286</link><guid>https://blog.youkuaiyun.com/WINCJS/article/details/105763286</guid><author>WINCJS</author><pubDate>Sun, 26 Apr 2020 10:19:51 +0800</pubDate><description><![CDATA[论文学习笔记MemGuard: Defending against Black-Box Membership Inference Attacks via Adversarial Examples

背景

•文章：Membership Inference Attacks Against Machine Learning Models. IEEE Symposium on Security and...]]></description><category></category></item><item><title><![CDATA[论文学习笔记 Helen: Maliciously Secure Coopetitive Learning for Linear Models]]></title><link>https://blog.youkuaiyun.com/WINCJS/article/details/103985227</link><guid>https://blog.youkuaiyun.com/WINCJS/article/details/103985227</guid><author>WINCJS</author><pubDate>Wed, 15 Jan 2020 11:02:15 +0800</pubDate><description><![CDATA[]]></description><category></category></item><item><title><![CDATA[论文学习笔记 Secure Two-party Threshold ECDSA from ECDSA Assumption]]></title><link>https://blog.youkuaiyun.com/WINCJS/article/details/103984746</link><guid>https://blog.youkuaiyun.com/WINCJS/article/details/103984746</guid><author>WINCJS</author><pubDate>Wed, 15 Jan 2020 10:51:32 +0800</pubDate><description><![CDATA[Secure Two-party Threshold ECDSA from ECDSA Assumptions 基于ECDSA假设的安全两方门限ECDSA1、问题引入2、问题改进The ECDSA AlgorithmTwo-party ECDSATwo-party Protocol2-of-n Protocol
论文引用：Doerner J , Kondi Y , Lee E , et al. S...]]></description><category></category></item></channel></rss>