search

OpenMined / PyDP

The Python Differential Privacy Library. Built on top of: https://github.com/google/differential-privacy
Apache License 2.0
507 stars 138 forks source link

Laplacian mechanism introduction #209

Closed chinmayshah99 closed 3 years ago

chinmayshah99 commented 4 years ago

Who?

Developer who's trying to understand what does it mean when you select laplacian

What?

Explanation of what choosing Laplacian mechanism for DP means and underlying assumption as well as recommendations on how noise is added and what type of data is desirable.

Additional Context

recommend working actively with research team

zzzuzum commented 4 years ago

Hi, I'll take this. Any specific place where this should be written?

chinmayshah99 commented 4 years ago

It needs to be put in the /docs folder. @Kritikalcoder seems to be already researching on it. Maybe reach out to her on slack?

NateSolon commented 4 years ago

I am working on a Jupyter notebook demo of the Laplacian mechanism, I'm hoping this can be a useful supplement to the documentation: https://github.com/OpenMined/writing/issues/110

replomancer commented 4 years ago

I read and executed your notebook @NateSolon Very clear and easy to follow! I think we should at least include a link to it somewhere in PyDP docs.

chinmayshah99 commented 3 years ago

307