paper-with-me

홈 › Papers

Privacy Amplification by Subsampling in Time Domain

2022-01-13 · Tatsuki Koga, Casey Meehan, Kamalika Chaudhuri

Aggregate time-series data like traffic flow and site occupancy repeatedly sample statistics from a population across time. Such data can be profoundly useful for understanding trends within a given population, but also pose a significant privacy risk, potentially revealing e.g., who spends time where. Producing a private version of a time-series satisfying the standard definition of Differential Privacy (DP) is challenging due to the large influence a single participant can have on the sequence: if an individual can contribute to each time step, the amount of additive noise needed to satisfy privacy increases linearly with the number of time steps sampled. As such, if a signal spans a long duration or is oversampled, an excessive amount of noise must be added, drowning out underlying trends. However, in many applications an individual realistically cannot participate at every time step. When this is the case, we observe that the influence of a single participant (sensitivity) can be reduced by subsampling and/or filtering in time, while still meeting privacy requirements. Using a novel analysis, we show this significant reduction in sensitivity and propose a corresponding class of privacy mechanisms. We demonstrate the utility benefits of these techniques empirically with real-world and synthetic time-series data.

📄 PDF Abstract BibTeX arXiv:2201.04762

Code (0)

등록된 구현이 없습니다.

Tasks

SensitivityTime SeriesTime Series Analysis

Similar Papers 제목 키워드 기반

Privacy Amplification by Subsampling: Tight Analyses via Couplings and Divergences

2018-07-04 · NeurIPS 2018 12 · Borja Balle, Gilles Barthe, Marco Gaboardi

Differential privacy comes equipped with multiple analytical tools for the design of private data analyses. One important tool is the so-called "privacy amplification by subsampling" principle, which ensures that a diffe…

Privacy Amplification by Structured Subsampling for Deep Differentially Private Time Series Forecasting

2025-02-04 · Jan Schuchardt, Mina Dalirrooyfard, Jed Guzelkabaagac, Anderson Schneider 외

Many forms of sensitive data, such as web traffic, mobility data, or hospital occupancy, are inherently sequential. The standard method for training machine learning models while ensuring privacy for units of sensitive i…

Data AugmentationTime SeriesTime Series Forecasting

Unified Mechanism-Specific Amplification by Subsampling and Group Privacy Amplification

2024-03-07 · Jan Schuchardt, Mihail Stoian, Arthur Kosmala, Stephan Günnemann

Amplification by subsampling is one of the main primitives in machine learning with differential privacy (DP): Training a model on random batches instead of complete datasets results in stronger privacy. This is traditio…

Composition of Differential Privacy & Privacy Amplification by Subsampling

2022-10-02 · Thomas Steinke

This chapter is meant to be part of the book "Differential Privacy for Artificial Intelligence Applications." We give an introduction to the most important property of differential privacy -- composition: running multipl…

Privacy Amplification via Random Participation in Federated Learning

2022-05-03 · Burak Hasircioglu, Deniz Gunduz

Running a randomized algorithm on a subsampled dataset instead of the entire dataset amplifies differential privacy guarantees. In this work, in a federated setting, we consider random participation of the clients in add…

Federated Learning