Abstract:Author(s): Song, Shuang | Advisor(s): Chaudhuri, Kamalika | Abstract: Modern machine learning increasingly involves personal data, such as healthcare, financial and user behavioral information. However, models trained on such data can reveal detailed information about the data and cause a serious privacy breach. Consequently, it is important to design algorithms that can analyze the sensitive data while still preserving privacy. This thesis advances the state-of-the-art of privacy-preserving machine learning in the following two major aspects.First, this thesis addresses the challenges in differentially private large-scale machine learning. On the one hand, with a large amount of sensitive user data, privacy-preserving learning algorithms are expected to achieve improved utility. On the other hand, big data imposes additional challenges, including performance (a.k.a Volume), data noise (a.k.a Veracity), a large number of classes and distributed sources (a.k.a Variety). This thesis presents (1) private versions of the widely used stochastic gradient descent (SGD) algorithm with generalization to data from multiple sources of different privacy requirements, and (2) an improved version of the Private Aggregation of Teacher Ensembles (PATE) framework which can scale to learning tasks with a large number of output classes and uncurated, imbalanced training data.Second, this thesis considers privacy-preserving data analysis beyond tabular data. Differential privacy is best suited for tabular data, where each record corresponds to all the information about an individual and records are independent of each other. However, many real-world applications involve non-tabular sensitive data, such as epidemic transmission graphs and measurement of the physical activity of a single subject across time. To analyze disease transmission graphs, or more general, graphs with sensitive information stored in each node, this thesis considers privacy-preserving continual release of graph statistics, such as the percentages of highly-active patients over time. The proposed algorithm outperforms the baselines in utility over a range of parameters. For physical activity measurement, or more generally, data with correlation, this thesis looks at a recent generalization of differential privacy, called Pufferfish privacy, that addresses privacy concerns in correlated data. Two mechanisms that work under different scenarios are proposed, and one of them is evaluated on real and synthetic time-series data.

Private Linear Regression with Differential Privacy and PAC Privacy

Differentially Private Simple Linear Regression

Differentially private regression analysis with dynamic privacy allocation

Differential Privacy Preservation in Regression Analysis Based on Relevance

Better Private Linear Regression Through Better Private Feature Selection

Differentially Private Regression with Unbounded Covariates

Differentially Private Linear Regression over Fully Decentralized Datasets

Differentially Private Model Selection with Penalized and Constrained Likelihood

Differentially Private Linear Regression Analysis via Truncating Technique

The Cost of Privacy: Optimal Rates of Convergence for Parameter Estimation with Differential Privacy

Privacy-Preserving Distributed Linear Regression on High-Dimensional Data

SoK: A Review of Differentially Private Linear Models For High-Dimensional Data

Privacy-Preserving Algorithms for Machine Learning

Evaluating Differentially Private Machine Learning in Practice

Privacy at a Price: Exploring its Dual Impact on AI Fairness

Impact of Prior Knowledge and Data Correlation on Privacy Leakage: A Unified Analysis.

Revisiting differentially private linear regression: optimal and adaptive prediction & estimation in unbounded domain

Differentially Private Random Feature Model

Analyzing the Differentially Private Theil-Sen Estimator for Simple Linear Regression

Improving the Privacy and Practicality of Objective Perturbation for Differentially Private Linear Learners

Chasing Your Long Tails: Differentially Private Prediction in Health Care Settings