Browsing Mathematics (Faculty of) by Subject "differential privacy"
Now showing items 1-12 of 12
-
Analyzing Threats of Large-Scale Machine Learning Systems
(University of Waterloo, 2024-02-22)Large-scale machine learning systems such as ChatGPT rapidly transform how we interact with and trust digital media. However, the emergence of such a powerful technology faces a dual-use dilemma. While it can have many ... -
Data Depth Inference for Difficult Data
(University of Waterloo, 2022-07-18)We explore various ways in which a robust, nonparametric statistical tool, the data depth function can be used to conduct inference on data which could be described as difficult. This can include data which are difficult ... -
Differential Privacy for Nearest Neighbor Queries
(University of Waterloo, 2022-08-16)We examine the problem of providing differential privacy for nearest neighbor queries. Very few mechanisms exist that achieve this, most notable geo-indistinguishability in the context of location privacy. However it uses ... -
Differentially Private Searchable Symmetric Encryption Scheme with Configurable Pattern Leakage
(University of Waterloo, 2019-12-19)Searchable symmetric encryption (SSE) allows a data owner to outsource its data to a cloud server while maintaining the ability to search over it. Most existing SSE schemes leak access-pattern leakage, and thus are vulnerable ... -
Differentially Private Simple Genetic Algorithms
(University of Waterloo, 2021-12-13)We study the differentially private (DP) selection problem, where the goal is to select an item from a set of candidates that approximately maximizes a given objective function. The most common solution to this problem is ... -
Differentially-private Multiparty Clustering
(University of Waterloo, 2023-09-13)In an era marked by the widespread application of Machine Learning (ML) across diverse domains, the necessity of privacy-preserving techniques has become paramount. The Euclidean k-Means problem, a fundamental component ... -
DP-Select: Improving Utility and Privacy in Tabular Data Synthesis with Differentially Private Generative Adversarial Networks and Differentially Private Selection
(University of Waterloo, 2023-05-29)This thesis proposes DP-Select, a novel approach to tabular data synthesis that combines DP-GAN and differentially private selection. We develop a mutual information-based selection method that is flexible and scalable for ... -
DProvSQL: Accuracy-Aware Privacy Provenance Framework for Differentially Private SQL Engine
(University of Waterloo, 2022-08-26)Recent years have witnessed the adoption of differential privacy (DP) in practical database query systems. Such systems, like PrivateSQL and FLEX, allow data analysts to query sensitive data while providing a rigorous and ... -
Efficient and Differentially Private Statistical Estimation via a Sum-of-Squares Exponential Mechanism
(University of Waterloo, 2023-01-30)As machine learning is applied to more privacy-sensitive data, it is becoming increasingly crucial to develop algorithms that maintain privacy. However, even the most basic high-dimensional statistical estimation tasks ... -
Private Distribution Learning with Public Data
(University of Waterloo, 2024-01-22)We study the problem of private distribution learning with access to public data. In this setup, a learner is given both public and private samples drawn from an unknown distribution 𝑝 belonging to a class 𝑄, and has the ... -
Trade-Offs between Fairness, Interpretability, and Privacy in Machine Learning
(University of Waterloo, 2020-05-14)Algorithms have increasingly been deployed to make consequential decisions, and there have been many ethical questions raised about how these algorithms function. Three ethical considerations we look at in this work are ... -
Unbiased Statistical Estimation and Valid Confidence Intervals Under Differential Privacy
(University of Waterloo, 2022-07-13)We present a method for producing unbiased parameter estimates and valid confidence intervals under the constraints of differential privacy, a formal framework for limiting individual information leakage from sensitive ...