Private Continual Release of Real-Valued Data Streams

Abstract

We present a differentially private mechanism to display statistics (e.g., the moving average) of a stream of real valued observations where the bound on each observation is either too conservative or unknown in advance. This is particularly relevant to scenarios of real-time data monitoring and reporting, e.g., energy data through smart meters. Our focus is on real-world data streams whose distribution is light-tailed, meaning that the tail approaches zero at least as fast as the exponential distribution. For such data streams, individual observations are expected to be concentrated below an unknown threshold. Estimating this threshold from the data can potentially violate privacy as it would reveal particular events tied to individuals [1]. On the other hand an overly conservative threshold may impact accuracy by adding more noise than necessary. We construct a utility optimizing differentially private mechanism to release this threshold based on the input stream. Our main advantage over the state-of-the-art algorithms is that the resulting noise added to each observation of the stream is scaled to the threshold instead of a possibly much larger bound; resulting in considerable gain in utility when the difference is significant. Using two real-world datasets, we demonstrate that our mechanism, on average, improves the utility by a factor of 3.5 on the first dataset, and 9 on the other. While our main focus is on continual release of statistics, our mechanism for releasing the threshold can be used in various other applications where a (privacy-preserving) measure of the scale of the input distribution is required.

Review of results on smart‐meter privacy by data manipulation, demand shaping, and load scheduling

“…[82] for more information on composition theorem. Recently, dynamic problems have been considered in differential privacy with improved privacy‐utility trade‐offs [146–149]. However, these studies mostly consider restrictive queries, such as counting.…”

“…They further require that the magnitude of the additive differential‐privacy noise to grow unboundedly, albeit at a much slower rate, e.g. logarithmically [148]. In fact, the magnitude of the additive differential‐privacy noise might need to grow boundedly for general queries [150].…”

Privacy-Driven Learning Analytics

“…It could for example preserve how single, pairs or n-tuples of attributes are distributed. Other contributions proposed DP-based mechanisms to allow interactive statistical queries about a dataset to be answered, again with some high accuracy on the results and some interesting properties, such as example supporting streams of data (Perrier et al, 2020). More recent work proposed DP-based additions to existing machine learning algorithms to allow the learning of models with strong privacy assurance.…”

Temporally Discounted Differential Privacy for Evolving Datasets on an Infinite Horizon

“…Recently, better performance bounds have been derived for differentially-private responses to queries on evolving datasets [7,8,9]. These studies however consider certain sets of queries, such as counting queries [7].…”