《【2】用Analytics-Zoo和Flink实现隐私保护的Cluster Serving-龚奇源-英特尔.pdf》由会员分享,可在线阅读,更多相关《【2】用Analytics-Zoo和Flink实现隐私保护的Cluster Serving-龚奇源-英特尔.pdf(36页珍藏版)》请在三个皮匠报告上搜索。
1、 ? ?6? ? 0 CA? ? ?29? ? ?/? ? ? ? ? ? ? ? 32 ? ?0? ?0?1?0? ? ? ? ? ? ? ? ? ? AIon Big Data *Other names and brands may be claimed as the property of others. ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? *Other names and brands may be claimed as the property of others. ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?
2、 ? ? ? ? ? ? ? *Other names and brands may be claimed as the property of others. ? ? ?人脸数据? ? ?人脸识别? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ?保护个人隐私的前提下?数据应用 ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? ? *Other names and brands may be claimed as the property of others. ? ? ? ? ? ? ?
3、? ? ?/?/?/? ? ?/? ? ? ? ? ? ? ? ? HEMPCTEE Clear Text Clear Text HEMPCTEE Security Performance Data Utility Clear Text HE DP DP DPMPCTEE TEE (Trust Execution Environment) HE (Homomorphic Encryption) DP (Differential privacy) MPC (Multi-party computation) Key takeaways TEE (e.g., SGX) performs well in all measures MPC & HEs performance is not good DP is widely used by Apple & Google, but it impacts