CRNov 4, 2020

The Limits of Differential Privacy (and its Misuse in Data Release and Machine Learning)

arXiv:2011.02352v1129 citations
Originality Synthesis-oriented
AI Analysis

It addresses the risk of overreliance on differential privacy for researchers and practitioners in data privacy, noting that this is an incremental critique rather than a new solution.

The paper reviews the limitations of differential privacy, highlighting that it is not a universal solution for all privacy problems and can be misused in contexts like individual data collection, release, and machine learning.

Differential privacy (DP) is a neat privacy definition that can co-exist with certain well-defined data uses in the context of interactive queries. However, DP is neither a silver bullet for all privacy problems nor a replacement for all previous privacy models. In fact, extreme care should be exercised when trying to extend its use beyond the setting it was designed for. This paper reviews the limitations of DP and its misuse for individual data collection, individual data release, and machine learning.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes