IBM Fully Homomorphic Encryption Toolkit For Linux. This toolkit is a Linux based Docker container that demonstrates computing on encrypted data without decrypting it! The toolkit ships with two demos including a fully encrypted Machine Learning inference with a Neural Network and a Privacy-Preserving key-value search.
IBM Linux全同态加密工具包。该工具包是一个基于Linux的Docker容器,它演示了在加密数据上进行计算,而无需解密数据!该工具包附带了两个演示,包括一个带有神经网络的完全加密机器学习推理和一个隐私保护键值搜索。
https://github.com/ibm/fhe-toolkit-linux
This FHE toolkit example demonstrates a privacy preserving search against an encrypted database. The database is a key value store prepopulated with the english names of countries and their capital cities. Selecting the country will use HElib to perform a search of the matching capital. BGV
refers to the encryption scheme used in the demo, more information on the schemes is available here.
FHE工具箱示例演示了针对加密数据库的隐私保护搜索。数据库是一个键值存储,预先填充了国家及其首都城市的英文名称。选择国家将使用HElib搜索匹配资本。BGV是指演示中使用的加密方案,有关这些方案的更多信息,请参阅此处。
https://github.com/IBM/fhe-toolkit-linux/tree/master/samples/BGV_world_…
示例目录包含教程和独立的示例程序,这些程序表达了各种API以及使用HElib的简单用例。
https://github.com/IBM-HElib/HElib/tree/master/examples
Microsoft SEAL是一个易于使用且功能强大的同态加密库。
https://github.com/Microsoft/SEAL
差异隐私验证器和运行时
https://github.com/opendp/smartnoise-core
什么是负责任的人工智能(预览)?
https://docs.microsoft.com/en-us/azure/machine-learning/concept-respons…
谷歌的差异隐私库
https://github.com/google/differential-privacy
差别隐私
该库包含用于生成数据集上ε-和(ε,δ)-差异私有统计的库。它包含以下工具。
Beam上的隐私是建立在Apache Beam之上的端到端差异隐私框架。它旨在易于使用,即使是非专家也可以使用。
C++、Go和Java中的三个“DP构建块”库。这些库实现基本噪声添加原语和差异私有聚合。使用这些库实现了Beam上的隐私。
随机测试仪,用于帮助捕获可能使差异隐私属性不再有效的回归。
差异隐私会计库,用于跟踪隐私预算。
用于使用ZetaSQL运行差异私有SQL查询的命令行界面。
要开始生成不同的私有数据,我们建议您遵循Beam codelab上的隐私。
目前,DP构建块库支持以下算法:
Algorithm | C++ | Go | Java |
---|---|---|---|
Laplace mechanism | Supported | Supported | Supported |
Gaussian mechanism | Supported | Supported | Supported |
Count | Supported | Supported | Supported |
Sum | Supported | Supported | Supported |
Mean | Supported | Supported | Supported |
Variance | Supported | Supported | Supported |
Standard deviation | Supported | Supported | Planned |
Quantiles | Supported | Supported | Supported |
Automatic bounds approximation | Supported | Planned | Supported |
Truncated geometric thresholding | Supported | Supported | Supported |
Laplace thresholding | Supported | Supported | Supported |
Gaussian thresholding | Planned | Supported | Supported |
- 登录 发表评论