Mehdi Amian

Calculating the privacy budget in the absence of additive noise

9

Hello. The "rdp_accountant" code assumes an additive Sampled Gaussian Mechanism (SGM). However, I am interested in calculating the privacy budget (epsilon and delta) when no noise is added. In this...

Following the original DPSGD algorithm (Abadi et al, 2016)

4

Hi. First off, I presume that the code, particularly "dp_optimizer.py" implements the original algorithm proposed by Abadai et al, 2016 (https://arxiv.org/abs/1607.00133). If this is not the case, correct me please....