search

OpenMined / PipelineDP

PipelineDP is a Python framework for applying differentially private aggregations to large datasets using batch processing systems such as Apache Spark, Apache Beam, and more.
https://pipelinedp.io/
Apache License 2.0
270 stars 75 forks source link

Add noise in explain computation report #432

Closed dvadym closed 1 year ago

dvadym commented 1 year ago

This PR extends explain computation report by adding noise kind and noise parameters for count/privacy_id_count/sum. For other aggregation it will be added later

Now it looks like

  1. Computed DP count with Gaussian mechanism: parameter=17.64452389929557 eps=0.3333333333333333 delta=3.333333333333333e-07 l2_sensitivity=1.4142135623730951
    1. Computed DP sum with Gaussian mechanism: parameter=88.22261949647785 eps=0.3333333333333333 delta=3.333333333333333e-07 l2_sensitivity=7.0710678118654755
    2. Computed DP privacy_id_count with Gaussian mechanism: parameter=17.64452389929557 eps=0.3333333333333333 delta=3.333333333333333e-07 l2_sensitivity=1.4142135623730951
dvadym commented 1 year ago

Thanks for review!