[{"data":1,"prerenderedAt":302},["ShallowReactive",2],{"content-query-811msiGuBp":3},{"_path":4,"_dir":5,"_draft":6,"_partial":6,"_locale":7,"title":8,"description":9,"date":10,"cover":11,"type":12,"body":13,"_type":296,"_id":297,"_source":298,"_file":299,"_stem":300,"_extension":301},"/news/zh/1489","zh",false,"","5月17日 Security SIG预告——密态计算技术分享，带你了解密码学在机器学习领域的创新应用","除了差分隐私，还有同态加密和安全多方计算等密态计算方法，它们是密码学在机器学习领域的创新应用，密态计算有着更高的安全保护强度。","2022-05-16","https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/05/24/35fb6d0018f947fc9685f0e0364e5547.png","news",{"type":14,"children":15,"toc":293},"root",[16,24,33,42,47,52,71,76,81,89,109,123,133,141,156,164,174,188,197,211,216,221,229,238,248,253,262,270,275,283,288],{"type":17,"tag":18,"props":19,"children":21},"element","h1",{"id":20},"_5月17日-security-sig预告密态计算技术分享带你了解密码学在机器学习领域的创新应用",[22],{"type":23,"value":8},"text",{"type":17,"tag":25,"props":26,"children":27},"p",{},[28],{"type":17,"tag":29,"props":30,"children":32},"img",{"alt":7,"src":31},"https://obs-mindspore-file.obs.cn-north-4.myhuaweicloud.com/file/2022/05/24/27ec5aa230344b4da1233bb12c7fcc37.gif",[],{"type":17,"tag":25,"props":34,"children":35},{},[36],{"type":17,"tag":37,"props":38,"children":39},"strong",{},[40],{"type":23,"value":41},"密态计算技术",{"type":17,"tag":25,"props":43,"children":44},{},[45],{"type":23,"value":46},"在之前的SIG会议中，我们揭示了联邦学习过程中传输明文信息（比如梯度信息或模型权重）存在的风险：攻击者可以基于这些明文信息，使用重构攻击或成员推理攻击去还原或推测部分原始数据信息，从而导致数据隐私的泄露。所以为了防止本地数据的隐私泄露，我们需要使用合适的安全策略对传输的明文信息进行保护。",{"type":17,"tag":25,"props":48,"children":49},{},[50],{"type":23,"value":51},"MindSpore Federated中实现了基于秘密分享的安全聚合算法和基于局部差分隐私的加噪方案以及SignDS方案，MindSpore Armour中实现了集中式训练的中心化差分隐私方案（DPSGD）。实际上，除了差分隐私，还有同态加密和安全多方计算等密态计算方法，它们是密码学在机器学习领域的创新应用，密态计算有着更高的安全保护强度。",{"type":17,"tag":25,"props":53,"children":54},{},[55,57,62,64,69],{"type":23,"value":56},"本次SIG会议我们邀请到了",{"type":17,"tag":37,"props":58,"children":59},{},[60],{"type":23,"value":61},"华为隐私计算专家****Zhou Lijing",{"type":23,"value":63},"和",{"type":17,"tag":37,"props":65,"children":66},{},[67],{"type":23,"value":68},"浙大研究生张朱谷承",{"type":23,"value":70},"和大家进行密态计算的技术交流。",{"type":17,"tag":25,"props":72,"children":73},{},[74],{"type":23,"value":75},"Zhou老师会给大家介绍业界密态计算的发展历史，并着重介绍安全多方计算、同态加密等技术以及密态计算面临的挑战。张朱谷承将为大家介绍纵向联邦必不可少的一个安全操作：隐私集合求交（PSI），同时会分享实现PSI的三种经典算法。",{"type":17,"tag":25,"props":77,"children":78},{},[79],{"type":23,"value":80},"会议最后会介绍开源实习任务，期待大家的参与~",{"type":17,"tag":25,"props":82,"children":83},{},[84],{"type":17,"tag":37,"props":85,"children":86},{},[87],{"type":23,"value":88},"1.1 Tips",{"type":17,"tag":90,"props":91,"children":92},"ul",{},[93,99,104],{"type":17,"tag":94,"props":95,"children":96},"li",{},[97],{"type":23,"value":98},"会议结束之后，我们会把视频上传至MindSpore的bilibili官方账号。",{"type":17,"tag":94,"props":100,"children":101},{},[102],{"type":23,"value":103},"此外，我们也非常欢迎AI安全领域（如对抗样本、联邦学习、差分隐私、安全多方计算、模型可解释、deepfake、语音仿冒）的老师同学们主动报名，和大家分享自己的成果，找到志同道合的伙伴，共同解决难题，共同成长！",{"type":17,"tag":94,"props":105,"children":106},{},[107],{"type":23,"value":108},"AI Security SIG代码仓：",{"type":17,"tag":25,"props":110,"children":111},{},[112,114],{"type":23,"value":113},"（1）",{"type":17,"tag":115,"props":116,"children":120},"a",{"href":117,"rel":118},"https://gitee.com/mindspore/mindarmour%EF%BC%882%EF%BC%89https://gitee.com/mindspore/mindspore/tree/master/mindspore/ccsrc/fl%EF%BC%88%E8%81%94%E9%82%A6%E5%AD%A6%E4%B9%A0%E4%BA%91%E4%BE%A7%EF%BC%89",[119],"nofollow",[121],{"type":23,"value":122},"https://gitee.com/mindspore/mindarmour（2）https://gitee.com/mindspore/mindspore/tree/master/mindspore/ccsrc/fl（联邦学习云侧）",{"type":17,"tag":25,"props":124,"children":125},{},[126],{"type":17,"tag":115,"props":127,"children":130},{"href":128,"rel":129},"https://gitee.com/mindspore/mindspore/tree/master/mindspore/lite/java/java/fl%5C_client%EF%BC%88%E8%81%94%E9%82%A6%E5%AD%A6%E4%B9%A0%E7%AB%AF%E4%BE%A7%EF%BC%89",[119],[131],{"type":23,"value":132},"https://gitee.com/mindspore/mindspore/tree/master/mindspore/lite/java/java/fl\\_client（联邦学习端侧）",{"type":17,"tag":25,"props":134,"children":135},{},[136],{"type":17,"tag":37,"props":137,"children":138},{},[139],{"type":23,"value":140},"1.2 会议详情",{"type":17,"tag":25,"props":142,"children":143},{},[144,149,151],{"type":17,"tag":37,"props":145,"children":146},{},[147],{"type":23,"value":148},"NO.1",{"type":23,"value":150}," ",{"type":17,"tag":37,"props":152,"children":153},{},[154],{"type":23,"value":155},"会议时间",{"type":17,"tag":25,"props":157,"children":158},{},[159],{"type":17,"tag":37,"props":160,"children":161},{},[162],{"type":23,"value":163},"2022/05/17（周二）",{"type":17,"tag":25,"props":165,"children":166},{},[167,172],{"type":17,"tag":37,"props":168,"children":169},{},[170],{"type":23,"value":171},"19:00-21:00",{"type":23,"value":173}," (GMT+08:00)",{"type":17,"tag":25,"props":175,"children":176},{},[177,182,183],{"type":17,"tag":37,"props":178,"children":179},{},[180],{"type":23,"value":181},"NO.2",{"type":23,"value":150},{"type":17,"tag":37,"props":184,"children":185},{},[186],{"type":23,"value":187},"会议链接",{"type":17,"tag":25,"props":189,"children":190},{},[191],{"type":17,"tag":115,"props":192,"children":195},{"href":193,"rel":194},"https://meeting.tencent.com/dm/Gi3p6McnqGcH",[119],[196],{"type":23,"value":193},{"type":17,"tag":25,"props":198,"children":199},{},[200,205,206],{"type":17,"tag":37,"props":201,"children":202},{},[203],{"type":23,"value":204},"NO.3",{"type":23,"value":150},{"type":17,"tag":37,"props":207,"children":208},{},[209],{"type":23,"value":210},"腾讯会议ID",{"type":17,"tag":25,"props":212,"children":213},{},[214],{"type":23,"value":215},"728-832-704",{"type":17,"tag":25,"props":217,"children":218},{},[219],{"type":23,"value":220},"（请大家在会议开始之前点击链接，根据链接提示下载好会议软件，谢谢！）",{"type":17,"tag":25,"props":222,"children":223},{},[224],{"type":17,"tag":37,"props":225,"children":226},{},[227],{"type":23,"value":228},"1.3 议程",{"type":17,"tag":25,"props":230,"children":231},{},[232,236],{"type":17,"tag":37,"props":233,"children":234},{},[235],{"type":23,"value":148},{"type":23,"value":237},"**《密态计算技术》**",{"type":17,"tag":25,"props":239,"children":240},{},[241,246],{"type":17,"tag":37,"props":242,"children":243},{},[244],{"type":23,"value":245},"时间",{"type":23,"value":247},"**：19:10 ~ 19:50**",{"type":17,"tag":25,"props":249,"children":250},{},[251],{"type":23,"value":252},"演讲人：Zhou Lijing，华为隐私计算专家",{"type":17,"tag":25,"props":254,"children":255},{},[256,260],{"type":17,"tag":37,"props":257,"children":258},{},[259],{"type":23,"value":181},{"type":23,"value":261},"**《隐私集合求交技术（PSI）及其应用》**",{"type":17,"tag":25,"props":263,"children":264},{},[265],{"type":17,"tag":37,"props":266,"children":267},{},[268],{"type":23,"value":269},"时间：20:00 ~ 20:40",{"type":17,"tag":25,"props":271,"children":272},{},[273],{"type":23,"value":274},"演讲人：张朱谷承，SIG成员-来自浙江大学网络空间安全学院",{"type":17,"tag":25,"props":276,"children":277},{},[278],{"type":17,"tag":37,"props":279,"children":280},{},[281],{"type":23,"value":282},"NO.3****开源实习任务发布",{"type":17,"tag":25,"props":284,"children":285},{},[286],{"type":23,"value":287},"时间：20:40 ~ 21:00",{"type":17,"tag":25,"props":289,"children":290},{},[291],{"type":23,"value":292},"实习导师介绍具体任务",{"title":7,"searchDepth":294,"depth":294,"links":295},4,[],"markdown","content:news:zh:1489.md","content","news/zh/1489.md","news/zh/1489","md",1776506049183]