We give a procedure for computing group-level $(\epsilon, \delta)$-DP guarantees for DP-SGD, when using Poisson sampling or fixed batch size sampling. Up to discretization errors in the implementation, the DP guarantees computed by this procedure are tight (assuming we release every intermediate iterate).

翻译：我们提出了一种计算DP-SGD在泊松采样或固定批量大小采样下的组级$(\epsilon, \delta)$-差分隐私保证的流程。在忽略实现中离散化误差的前提下，该流程所计算的差分隐私保证是紧致的（假设我们发布所有中间迭代结果）。