Impact of Social Learning on Privacy-Preserving Data Collection
Impact of Social Learning on Privacy-Preserving Data Collection
复制标题
DOI：
10.1109/jsait.2021.3053545
复制
发表时间：
2019-03
期刊：
影响因子：
--
通讯作者：
Abdullah Basar Akbay;Weina Wang;Junshan Zhang
中科院分区：
文献类型：
--
作者：
Abdullah Basar Akbay;Weina Wang;Junshan Zhang
我们研究了一个博弈论模型，其中数据收集器通过付费机制从用户那里购买数据。每个用户都有自己的个人信号，这代表了她对数据收集器想要了解的底层状态的了解。通过社交互动，每个用户还可以了解到她朋友的个人信号的嘈杂版本，这被称为“群信号”。我们建立了贝叶斯博弈论框架来研究社会学习对用户数据报告策略的影响，并相应地设计了数据收集者的付费机制。我们证明了贝叶斯-纳什均衡可以是对称随机响应（SR）策略或信息非披露（ND）策略。具体而言，每个用户对其嘈杂的群体信号应用广义多数投票规则来确定遵循哪种策略。我们的研究结果表明，数据收集者和用户都可以从社会学习中受益，这降低了隐私成本，并有助于改善给定总支付预算的状态估计。此外，我们推导了达到给定状态估计精度所需的最小总支付的界限。
We study a game-theoretic model where a data collector purchases data from users through a payment mechanism. Each user has her personal signal which represents her knowledge about the underlying state the data collector desires to learn. Through social interactions, each user can also learn noisy versions of her friends’ personal signals, which are called ‘group signals’. We develop a Bayesian game theoretic framework to study the impact of social learning on users’ data reporting strategies and devise the payment mechanism for the data collector accordingly. We show that the Bayesian-Nash equilibrium can be in the form of either a symmetric randomized response (SR) strategy or an informative non-disclosive (ND) strategy. Specifically, a generalized majority voting rule is applied by each user to her noisy group signals to determine which strategy to follow. Our findings reveal that both the data collector and the users can benefit from social learning which drives down the privacy costs and helps to improve the state estimation for a given total payment budget. Further, we derive bounds on the minimum total payment required to achieve a given level of state estimation accuracy.