This project contains a set of libraries of ε- and (ε, δ)-differentially private algorithms, which can be used to produce aggregate statistics over numeric data sets containing private or sensitive information. The functionality is currently available in C++ and Java.

Currently, we provide algorithms to compute the following:

We also provide an implementation of the Laplace and Gaussian mechanism that can be used to perform computations that aren't covered by our pre-built algorithms.

All of these algorithms are suitable for research, experimental or production use cases.

This project also contains a stochastic tester, used to help catch regressions that could make the differential privacy property no longer hold.

In order to run the differential private library, you need to install Bazel, if you don't have it already. Follow the instructions for your platform on the Bazel website

You also need to install Git, if you don't have it already. Follow the instructions for your platform on the Git website.

Once you've installed Bazel and Git, open a Terminal and clone the differential privacy directory into a local folder:

git clone https://github.com/google/differential-privacy.git

Navigate into the differential-privacy folder you just created, and build the differential privacy library and dependencies using Bazel:

To build the C++ library, run: bazel build differential_privacy/...

To build the Java library, run:

cd java
bazel build ...

You may need to install additional dependencies when building the PostgreSQL extension, for example on Ubuntu you will need these packages:

sudo apt-get install libreadline-dev bison flex

This is experimental and we are working on few TODOs to drop the build time.

To build the base docker image

./run_docker_build.sh

To run the example

./run_dockerized_example.sh

To run the tests

./run_dockerized_tests.sh

Full documentation on how to use the library is in the cpp/docs subdirectory. Here's a minimal example showing how to compute the count of some data:

Differential Privacy requires some bound on maximum number of contributions each user can make to a single partition. The libraries don't perform such bounding.

The libraries implementation assumes that each user contributes only a single row to each partition. It neither verifies nor enforces this; it is still the caller's responsibility to pre-process data to enforce this bound.

We chose not to implement this step at the library level because it's not the logical place for it - it's much easier to sort contributions by user and combine them together with a distributed processing framework before they're passed to our algorithms. You can use the library to build systems that allow multiple contributions per user - our paper describes one such system. To do so, multiple user contributions should be combined before they are passed to our algorithms.

We will continue to publish updates and improvements to the library. We will not accept pull requests for the immediate future. We will respond to issues filed in this project. If we intend to stop publishing improvements and responding to issues we will publish notice here at least 3 months in advance.

Apache License 2.0

This is not an officially supported Google product.