Learning from continuous data streams via classification/regression is prevalent in many domains. Adapting to evolving data characteristics (concept drift) while protecting data owners' private information is an open challenge. We present a differentially private ensemble solution to this problem with two distinguishing features: it allows an \textit{unbounded} number of ensemble updates to deal with the potentially never-ending data streams under a fixed privacy budget, and it is \textit{model agnostic}, in that it treats any pre-trained differentially private classification/regression model as a black-box. Our method outperforms competitors on real-world and simulated datasets for varying settings of privacy, concept drift, and data distribution.
翻译:通过分类/递减从连续数据流中学习,在许多领域很普遍。适应不断演变的数据特征(概念流),同时保护数据所有者的私人信息,这是一个公开的挑战。我们为这一问题提出了一个差别化的私人共同解决办法,有两个显著特点:它允许在固定隐私预算下进行数量不等的组合式更新,以处理可能永无止境的数据流,而它是 kextit{模型的不可知性},因为它将任何事先经过训练的有差别的私人分类/递增模型作为黑盒处理。我们的方法超越了现实世界的竞争者,并模拟数据集,用于不同的隐私、概念流和数据分布环境。