English

Multi-User Privacy Mechanism Design with Non-zero Leakage

Information Theory 2022-11-29 v1 math.IT

Abstract

A privacy mechanism design problem is studied through the lens of information theory. In this work, an agent observes useful data Y=(Y1,...,YN)Y=(Y_1,...,Y_N) that is correlated with private data X=(X1,...,XN)X=(X_1,...,X_N) which is assumed to be also accessible by the agent. Here, we consider KK users where user ii demands a sub-vector of YY, denoted by CiC_{i}. The agent wishes to disclose CiC_{i} to user ii. Since CiC_{i} is correlated with XX it can not be disclosed directly. A privacy mechanism is designed to generate disclosed data UU which maximizes a linear combinations of the users utilities while satisfying a bounded privacy constraint in terms of mutual information. In a similar work it has been assumed that XiX_i is a deterministic function of YiY_i, however in this work we let XiX_i and YiY_i be arbitrarily correlated. First, an upper bound on the privacy-utility trade-off is obtained by using a specific transformation, Functional Representation Lemma and Strong Functional Representaion Lemma, then we show that the upper bound can be decomposed into NN parallel problems. Next, lower bounds on privacy-utility trade-off are derived using Functional Representation Lemma and Strong Functional Representaion Lemma. The upper bound is tight within a constant and the lower bounds assert that the disclosed data is independent of all {Xj}i=1N\{X_j\}_{i=1}^N except one which we allocate the maximum allowed leakage to it. Finally, the obtained bounds are studied in special cases.

Keywords

Cite

@article{arxiv.2211.15525,
  title  = {Multi-User Privacy Mechanism Design with Non-zero Leakage},
  author = {Amirreza Zamani and Tobias J. Oechtering and Mikael Skoglund},
  journal= {arXiv preprint arXiv:2211.15525},
  year   = {2022}
}

Comments

arXiv admin note: text overlap with arXiv:2205.04881, arXiv:2201.08738

R2 v1 2026-06-28T07:15:16.482Z