arXiv:2202.12058

Exploring the Unfairness of DP-SGD Across Settings

Frederik Noe, Rasmus Herskind, Anders Søgaard

제출 2022-02-24

원문 초록

End users and regulators require private and fair artificial intelligence models, but previous work suggests these objectives may be at odds. We use the CivilComments to evaluate the impact of applying the {\em de facto} standard approach to privacy, DP-SGD, across several fairness metrics. We evaluate three implementations of DP-SGD: for dimensionality reduction (PCA), linear classification (logistic regression), and robust deep learning (Group-DRO). We establish a negative, logarithmic correlation between privacy and fairness in the case of linear classification and robust deep learning. DP-SGD had no significant impact on fairness for PCA, but upon inspection, also did not seem to lead to private representations.

함께 살펴볼 논문

공통 태그를 근거로 연결했습니다. 실제 인용 관계나 논문의 우수성을 뜻하지 않습니다.

논문 ID: document/9197689공통 태그 Dimensionality Reduction, Classification

Fuzzy Pooling

토론

이 논문과 연결된 포럼 글이 아직 없습니다.