Motivated by problems arising in digital advertising, we introduce the task
of training differentially private (DP) machine learning models with
semi-sensitive features. In this setting, a subset of the features is known to
the attacker (and thus need not be protected) while the remaining features as
well as the label are unknown to the attacker and should be protected by the DP
guarantee. This task interpolates between training the model with full DP
(where the label and all features should be protected) or with label DP (where
all the features are considered known, and only the label should be protected).
We present a new algorithm for training DP models with semi-sensitive features.
Through an empirical evaluation on real ads datasets, we demonstrate that our
algorithm surpasses in utility the baselines of (i) DP stochastic gradient
descent (DP-SGD) run on all features (known and unknown), and (ii) a label DP
algorithm run only on the known features (while discarding the unknown ones).

在数字广告中遇到的问题的基础上，我们引入了使用半敏感特征训练差分隐私（DP）机器学习模型的任务。在这个设置中，攻击者已知一部分特征（因此无需保护），而其余特征以及标签对攻击者来说是未知的，需要受到 DP 保护。该任务在使用完全 DP（需要保护标签和所有特征）或使用标签 DP（认为所有特征已知，只需保护标签）两者之间插值。我们提出了一种新的算法来训练具有半敏感特征的 DP 模型。通过对真实广告数据集进行实证评估，我们证明了我们的算法在效用上超过了（i）在所有特征（已知和未知）上运行的 DP 随机梯度下降（DP-SGD）基线，以及（ii）仅在已知特征上运行的标签 DP 算法（而舍弃未知特征）的基线。