The purpose of this paper is to guide interpretation of the semantic privacy guarantees for some of the major variations of differential privacy, which include pure, approximate, Rényi, zero-concentrated, and $f$ differential privacy. We interpret privacy-loss accounting parameters, frequentist semantics, and Bayesian semantics (including new results). The driving application is the interpretation of the confidentiality protections for the 2020 Census Public Law 94-171 Redistricting Data Summary File released August 12, 2021, which, for the first time, were produced with formal privacy guarantees.
The paper aims to give an overview of various approaches to statistical disclosure control based on random noise that are currently being discussed for official population statistics and censuses. A particular focus is on a stringent delineation between different concepts influencing the discussion: we separate clearly between risk measures, noise distributions and output mechanisms - putting these concepts into scope and into relation with each other. After recapitulating differential privacy as a risk measure, the paper also remarks on utility and risk aspects of some specific output mechani…
The Census TopDown Algorithm (TDA) is a disclosure avoidance system using differential privacy for privacy-loss accounting. The algorithm ingests the final, edited version of the 2020 Census data and the final tabulation geographic definitions. The algorithm then creates noisy versions of key queries on the data, referred to as measurements, using zero-Concentrated Differential Privacy. Another key aspect of the TDA are invariants, statistics that the Census Bureau has determined, as matter of policy, to exclude from the privacy-loss accounting. The TDA post-processes the measurements together…
The US Census Bureau will implement a new privacy-preserving disclosure avoidance system (DAS), which includes application of differential privacy, on the public-release 2020 census data. There are concerns that the DAS may bias small-area and demographically-stratified population counts, which play a critical role in public health research and policy, serving as denominators in estimation of disease/mortality rates. Employing three DAS demonstration products, we quantify errors attributable to reliance on DAS-protected denominators in standard small-area disease mapping models for characteriz…
This article describes a proposed differentially private (DP) algorithms that the US Census Bureau is considering to release the Detailed Demographic and Housing Characteristics (DHC) Race & Ethnicity tabulations as part of the 2020 Census. The tabulations contain statistics (counts) of demographic and housing characteristics of the entire population of the US crossed with detailed races and tribes at varying levels of geography. We describe two differentially private algorithmic strategies, one based on adding noise drawn from a two-sided Geometric distribution that satisfies "pure"-DP, and a…
We propose Bayesian methods to assess the statistical disclosure risk of data released under zero-concentrated differential privacy, focusing on settings with a strong hierarchical structure and categorical variables with many levels. Risk assessment is performed by hypothesizing Bayesian intruders with various amounts of prior information and examining the distance between their posteriors and priors. We discuss applications of these risk assessment methods to differentially private data releases from the 2020 decennial census and perform simulation studies using public individual-level data …
The US Census Bureau plans to protect the privacy of 2020 Census respondents through its Disclosure Avoidance System (DAS), which attempts to achieve differential privacy guarantees by adding noise to the Census microdata. By applying redistricting simulation and analysis methods to DAS-protected 2010 Census data, we find that the protected data are not of sufficient quality for redistricting purposes. We demonstrate that the injected noise makes it impossible for states to accurately comply with the One Person, One Vote principle. Our analysis finds that the DAS-protected data are biased agai…
Carlson School of Management, University of Minnesota（明尼苏达大学卡尔森管理学院）；School of Statistics, University of Minnesota（明尼苏达大学统计学院）
Differential privacy is becoming one gold standard for protecting the privacy of publicly shared data. It has been widely used in social science, data science, public health, information technology, and the U.S. decennial census. Nevertheless, to guarantee differential privacy, existing methods may unavoidably alter the conclusion of the original data analysis, as privatization often changes the sample distribution. This phenomenon is known as the trade-off between privacy protection and statistical accuracy. In this work, we mitigate this trade-off by developing a distribution-invariant priva…
We propose new differential privacy solutions for when external \emph{invariants} and \emph{integer} constraints are simultaneously enforced on the data product. These requirements arise in real world applications of private data curation, including the public release of the 2020 U.S. Decennial Census. They pose a great challenge to the production of provably private data products with adequate statistical usability. We propose \emph{integer subspace differential privacy} to rigorously articulate the privacy guarantee when data products maintain both the invariants and integer characteristics,…
In an era where external data and computational capabilities far exceed statistical agencies' own resources and capabilities, they face the renewed challenge of protecting the confidentiality of underlying microdata when publishing statistics in very granular form and ensuring that these granular data are used for statistical purposes only. Conventional statistical disclosure limitation methods are too fragile to address this new challenge. This article discusses the deployment of a differential privacy framework for the 2020 US Census that was customized to protect confidentiality, particular…
Many data applications have certain invariant constraints due to practical needs. Data curators who employ differential privacy need to respect such constraints on the sanitized data product as a primary utility requirement. Invariants challenge the formulation, implementation, and interpretation of privacy guarantees. We propose subspace differential privacy, to honestly characterize the dependence of the sanitized output on confidential aspects of the data. We discuss two design frameworks that convert well-known differentially private mechanisms, such as the Gaussian and the Laplace mechani…
Privacy and nondiscrimination are related but different. We make this observation precise in two ways. First, we show that both privacy and nondiscrimination have two versions, a causal version and a statical associative version, with each version corresponding to a competing view of the proper goal of privacy or nondiscrimination. Second, for each version, we show that a difference between the privacy edition of the version and the nondiscrimination edition of the version is related to the difference between Bayesian probabilities and frequentist probabilities. In particular, privacy admits b…
In machine learning, boosting is one of the most popular methods that designed to combine multiple base learners to a superior one. The well-known Boosted Decision Tree classifier, has been widely adopted in many areas. In the big data era, the data held by individual and entities, like personal images, browsing history and census information, are more likely to contain sensitive information. The privacy concern raises when such data leaves the hand of the owners and be further explored or mined. Such privacy issue demands that the machine learning algorithm should be privacy aware. Recently, …
Statistical modelling under differential privacy constraints: A case study in fine-scale geographical analysis with Australian Bureau of Statistics TableBuilder data
In 2018, the US Census Bureau designed a new data reconstruction and re-identification attack and tested it against their 2010 data release. The specific attack executed by the Bureau allows an attacker to infer the race and ethnicity of respondents with average 75% precision for 85% of the respondents, assuming that the attacker knows the correct age, sex, and address of the respondents. They interpreted the attack as exceeding the Bureau's privacy standards, and so introduced stronger privacy protections for the 2020 Census in the form of the TopDown Algorithm (TDA). This paper demonstrates …
CDC WONDER is a web-based tool for the dissemination of epidemiologic data collected by the National Vital Statistics System. While CDC WONDER has built-in privacy protections, they do not satisfy formal privacy protections such as differential privacy and thus are susceptible to targeted attacks. Given the importance of making high-quality public health data publicly available while preserving the privacy of the underlying data subjects, we aim to improve the utility of a recently developed approach for generating Poisson-distributed, differentially private synthetic data by using publicly av…
The Complexities of Differential Privacy for Survey Data
Institute for Employment Research（就业研究所）；Ludwig-Maximilians-Universität（路德维希-马克西米利安大学）；University of Maryland, College Park（马里兰大学帕克分校）；Harvard University（哈佛大学）
Syracuse University（雪城大学）；Georgia Institute of Technology（佐治亚理工学院）
This paper is motivated by applications of a Census Bureau interested in releasing aggregate socio-economic data about a large population without revealing sensitive information about any individual. The released information can be the number of individuals living alone, the number of cars they own, or their salary brackets. Recent events have identified some of the privacy challenges faced by these organizations. To address them, this paper presents a novel differential-privacy mechanism for releasing hierarchical counts of individuals. The counts are reported at multiple granularities (e.g.,…
The U.S. Census Bureau is using differential privacy (DP) to protect confidential respondent data collected for the 2020 Decennial Census of Population & Housing. The Census Bureau's DP system is implemented in the Disclosure Avoidance System (DAS) and requires a source of random numbers. We estimate that the 2020 Census will require roughly 90TB of random bytes to protect the person and household tables. Although there are critical differences between cryptography and DP, they have similar requirements for randomness. We review the history of random number generation on deterministic computer…
In this short paper, we outline the design of Tumult Analytics, a Python framework for differential privacy used at institutions such as the U.S. Census Bureau, the Wikimedia Foundation, or the Internal Revenue Service.
Georgia Institute of Technology（佐治亚理工学院）；Syracuse University（雪城大学）
Post-processing immunity is a fundamental property of differential privacy: it enables the application of arbitrary data-independent transformations to the results of differentially private outputs without affecting their privacy guarantees. When query outputs must satisfy domain constraints, post-processing can be used to project the privacy-preserving outputs onto the feasible region. Moreover, when the feasible region is convex, a widely adopted class of post-processing steps is also guaranteed to improve accuracy. Post-processing has been applied successfully in many applications including…
Georgia Institute of Technology（佐治亚理工学院）；Syracuse University（雪城大学）
Post-processing immunity is a fundamental property of differential privacy: it enables arbitrary data-independent transformations to differentially private outputs without affecting their privacy guarantees. Post-processing is routinely applied in data-release applications, including census data, which are then used to make allocations with substantial societal impacts. This paper shows that post-processing causes disparate impacts on individuals or groups and analyzes two critical settings: the release of differentially private datasets and the use of such private datasets for downstream deci…
When differential privacy was created more than a decade ago, the motivating example was statistics published by an official statistics agency. In attempting to transition differential privacy from the academy to practice, the U.S. Census Bureau has encountered many challenges unanticipated by differential privacy's creators. These challenges include obtaining qualified personnel and a suitable computing environment, the difficulty accounting for all uses of the confidential data, the lack of release mechanisms that align with the needs of data users, the expectation on the part of data users …
Applying differential privacy at scale requires convenient ways to check that programs computing with sensitive data appropriately preserve privacy. We propose here a fully automated framework for {\em testing} differential privacy, adapting a well-known "pointwise" technique from informal proofs of differential privacy. Our framework, called DPCheck, requires no programmer annotations, handles all previously verified or tested algorithms, and is the first fully automated framework to distinguish correct and buggy implementations of PrivTree, a probabilistically terminating algorithm that has …
Georgia Institute of Technology（佐治亚理工学院）；Syracuse University（雪城大学）；National Institute of Science Education and Research（国家科学教育与研究所）；Knexus Research Corporation（Knexus研究公司）
Disclosure avoidance (DA) systems are used to safeguard the confidentiality of data while allowing it to be analyzed and disseminated for analytic purposes. These methods, e.g., cell suppression, swapping, and k-anonymity, are commonly applied and may have significant societal and economic implications. However, a formal analysis of their privacy and bias guarantees has been lacking. This paper presents a framework that addresses this gap: it proposes differentially private versions of these mechanisms and derives their privacy bounds. In addition, the paper compares their performance with tra…
Differential privacy (DP) has arisen as the state-of-the-art metric for quantifying individual privacy when sensitive data are analyzed, and it is starting to see practical deployment in organizations such as the US Census Bureau, Apple, Google, etc. There are two popular models for deploying differential privacy - standard differential privacy (SDP), where a trusted server aggregates all the data and runs the DP mechanisms, and local differential privacy (LDP), where each user perturbs their own data and perturbed data is analyzed. Due to security concerns arising from aggregating raw data at…
Differential privacy offers a formal framework for reasoning about privacy and accuracy of computations on private data. It also offers a rich set of building blocks for constructing data analyses. When carefully calibrated, these analyses simultaneously guarantee privacy of the individuals contributing their data, and accuracy of their results for inferring useful properties about the population. The compositional nature of differential privacy has motivated the design and implementation of several programming languages aimed at helping a data analyst in programming differentially private ana…
Large organizations that collect data about populations (like the US Census Bureau) release summary statistics that are used by multiple stakeholders for resource allocation and policy making problems. These organizations are also legally required to protect the privacy of individuals from whom they collect data. Differential Privacy (DP) provides a solution to release useful summary data while preserving privacy. Most DP mechanisms are designed to answer a single set of queries. In reality, there are often multiple stakeholders that use a given data release and have overlapping but not-identi…
Local differential privacy has become the gold-standard of privacy literature for gathering or releasing sensitive individual data points in a privacy-preserving manner. However, locally differential data can twist the probability density of the data because of the additive noise used to ensure privacy. In fact, the density of privacy-preserving data (no matter how many samples we gather) is always flatter in comparison with the density function of the original data points due to convolution with privacy-preserving noise density function. The effect is especially more pronounced when using slo…
