opendp/opendp

The core library of differential privacy algorithms powering the OpenDP Project.

What it solves

OpenDP is designed to solve the problem of extracting statistical insights from data while protecting the privacy of individuals within that data. It provides a modular collection of algorithms that adhere to the mathematical definition of differential privacy, allowing developers to build applications that perform privacy-preserving computations without compromising sensitive information.

How it works

The library is implemented in Rust for performance and safety, with bindings for Python and R to make it accessible to data scientists and researchers. It uses a conceptual framework for expressing privacy-aware computations, enabling the modular assembly of statistical algorithms to create a privacy-preserving pipeline.

Who it’s for

It is intended for developers and data scientists who need to build privacy-critical applications or prototype privacy-preserving statistical analysis tools.

Highlights

  • own a modular collection of statistical algorithms based on differential privacy
  • Supports multiple programming languages via Rust, Python, and R bindings
  • Based on a formal programming framework for privacy-aware computations
  • Open-source under the MIT license

Related

  • Project
    google/differential-privacy

    A collection of libraries and frameworks for generating differentially private statistics over datasets to protect individual privacy in aggregate data analysis.

  • Project
    meta-pytorch/opacus

    A PyTorch library for training machine learning models with differential privacy, allowing users to protect sensitive training data while tracking privacy loss.

  • Project
    sdv-dev/SDV

    A Python library for creating tabular synthetic data using machine learning to emulate real-world patterns while maintaining privacy.

  • Project
    mostly-ai/mostlyai

    A Python toolkit for creating high-fidelity, privacy-safe synthetic tabular and language data to enable secure data sharing and analysis.

  • Project
    geomstats/geomstats

    A Python package for computations, statistics, and machine learning on manifolds, enabling the analysis of non-Euclidean data like rotations and shapes.