From 8923cc5385508b92b497a5beb5d0e3df656a3d0b Mon Sep 17 00:00:00 2001 From: Yuchen Pei Date: Thu, 14 Mar 2019 10:50:02 +0100 Subject: minor --- posts/2019-03-13-a-tail-of-two-densities.md | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-) (limited to 'posts/2019-03-13-a-tail-of-two-densities.md') diff --git a/posts/2019-03-13-a-tail-of-two-densities.md b/posts/2019-03-13-a-tail-of-two-densities.md index 98cedd5..dda906b 100644 --- a/posts/2019-03-13-a-tail-of-two-densities.md +++ b/posts/2019-03-13-a-tail-of-two-densities.md @@ -35,10 +35,13 @@ I also attempt to reproduce a seemingly better moment bound for the Gaussian mechanism with subsampling, with one intermediate step which I am not able to prove. -Finally I explain the Tensorflow implementation of differential privacy, -and using the results from both Part 1 and Part 2 to obtain some privacy -guarantees for the differentially private stochastic gradient descent -algorithm (DP-SGD). I also compare these privacy guarantees. +After that I explain the Tensorflow implementation of differential privacy, +which focuses on the differentially private stochastic gradient descent +algorithm (DP-SGD). + +Finally I use the results from both Part 1 and Part 2 to obtain some privacy +guarantees for composed subsampling queries in general, and for DP-SGD in particular. +I also compare these privacy guarantees. **Acknowledgement**. I would like to thank [Stockholm AI](https://stockholm.ai) for introducing me to the subject -- cgit v1.2.3