Privacy Amplification via Shuffling for Linear Contextual Bandits
Privacy Amplification via Shuffling for Linear Contextual Bandits
复制标题
通过线性上下文强盗的洗牌来增强隐私
DOI：
--
复制
发表时间：
2021
期刊：
影响因子：
--
通讯作者：
Matteo Pirotta
中科院分区：
文献类型：
--
作者：
Evrard Garcelon;Kamalika Chaudhuri;Vianney Perchet;Matteo Pirotta
上下文强盗算法广泛应用于希望通过利用上下文信息提供个性化服务的领域，其中可能包含需要保护的敏感信息。受此启发，我们研究了具有差分隐私（DP）约束的上下文线性盗匪问题。虽然文献关注的是集中式（联合DP）或本地（本地DP）隐私，但我们考虑了隐私的洗牌模型，并表明在JDP和LDP之间实现隐私/效用权衡是可能的。通过利用隐私的洗选和盗号的批处理，我们提出了一种具有遗憾绑定$ widdetilde {mathcal{O}}(T^{2/3}/varepsilon^{1/3})$的算法，同时保证了中心（联合）和局部隐私。我们的结果表明，在保留本地隐私的同时，利用shuffle模型可以在JDP和LDP之间获得权衡。
Contextual bandit algorithms are widely used in domains where it is desirable to provide a personalized service by leveraging contextual information, that may contain sensitive information that needs to be protected. Inspired by this scenario, we study the contextual linear bandit problem with differential privacy (DP) constraints. While the literature has focused on either centralized (joint DP) or local (local DP) privacy, we consider the shuffle model of privacy and we show that is possible to achieve a privacy/utility trade-off between JDP and LDP. By leveraging shuffling from privacy and batching from bandits, we present an algorithm with regret bound $widetilde{mathcal{O}}(T^{2/3}/varepsilon^{1/3})$, while guaranteeing both central (joint) and local privacy. Our result shows that it is possible to obtain a trade-off between JDP and LDP by leveraging the shuffle model while preserving local privacy.