Differentially private stochastic gradient descent (DP-SGD) is the workhorse algorithm for recent advances in private deep learning. It provides a single privacy guarantee to all datapoints in the dataset. We propose output-specific $(\varepsilon,\delta)$-DP to characterize privacy guarantees for individual examples when releasing models trained by DP-SGD. We also design an efficient algorithm to investigate individual privacy across a number of datasets. We find that most examples enjoy stronger privacy guarantees than the worst-case bound. We further discover that the training loss and the privacy parameter of an example are well-correlated. This implies groups that are underserved in terms of model utility simultaneously experience weaker privacy guarantees. For example, on CIFAR-10, the average $\varepsilon$ of the class with the lowest test accuracy is 44.2\% higher than that of the class with the highest accuracy.

翻译：差分隐私随机梯度下降法（DP-SGD）是近期深度隐私学习领域的基础算法。该算法为数据集中的所有数据点提供统一的隐私保障。本文提出输出特定的$(\varepsilon,\delta)$-DP概念，用于刻画训练DP-SGD模型时单个样本的隐私保障程度。我们同时设计了高效算法，在多个数据集上研究个体隐私差异。研究发现大多数样本享受比最坏情况更优的隐私保障。进一步发现，训练损失与样本隐私参数存在显著相关性：模型效用不足的群体同时面临更弱的隐私保障。例如在CIFAR-10数据集中，测试准确率最低类别样本的平均$\varepsilon$值较最高准确率类别高出44.2%。