skip to main content

Title: Privacy for All: Ensuring Fair and Equitable Privacy Protections
In this position paper, we argue for applying recent research on ensuring sociotechnical systems are fair and non-discriminatory to the privacy protections those systems may provide. Privacy literature seldom considers whether a proposed privacy scheme protects all persons uniformly, irrespective of membership in protected classes or particular risk in the face of privacy failure. Just as algorithmic decision-making systems may have discriminatory outcomes even without explicit or deliberate discrimination, so also privacy regimes may disproportionately fail to protect vulnerable members of their target population, resulting in disparate impact with respect to the effectiveness of privacy protections.We propose a research agenda that will illuminate this issue, along with related issues in the intersection of fairness and privacy, and present case studies that show how the outcomes of this research may change existing privacy and fairness research. We believe it is important to ensure that technologies and policies intended to protect the users and subjects of information systems provide such protection in an equitable fashion.
Authors:
; ;
Award ID(s):
1657774
Publication Date:
NSF-PAR ID:
10222637
Journal Name:
1st Conference on Fairness, Accountability and Transparency
Sponsoring Org:
National Science Foundation
More Like this
  1. Data sets and statistics about groups of individuals are increasingly collected and released, feeding many optimization and learning algorithms. In many cases, the released data contain sensitive information whose privacy is strictly regulated. For example, in the U.S., the census data is regulated under Title 13, which requires that no individual be identified from any data released by the Census Bureau. In Europe, data release is regulated according to the General Data Protection Regulation, which addresses the control and transfer of personal data. Differential privacy has emerged as the de-facto standard to protect data privacy. In a nutshell, differentially privatemore »algorithms protect an individual’s data by injecting random noise into the output of a computation that involves such data. While this process ensures privacy, it also impacts the quality of data analysis, and, when private data sets are used as inputs to complex machine learning or optimization tasks, they may produce results that are fundamentally different from those obtained on the original data and even rise unintended bias and fairness concerns. In this talk, I will first focus on the challenge of releasing privacy-preserving data sets for complex data analysis tasks. I will introduce the notion of Constrained-based Differential Privacy (C-DP), which allows casting the data release problem to an optimization problem whose goal is to preserve the salient features of the original data. I will review several applications of C-DP in the context of very large hierarchical census data, data streams, energy systems, and in the design of federated data-sharing protocols. Next, I will discuss how errors induced by differential privacy algorithms may propagate within a decision problem causing biases and fairness issues. This is particularly important as privacy-preserving data is often used for critical decision processes, including the allocation of funds and benefits to states and jurisdictions, which ideally should be fair and unbiased. Finally, I will conclude with a roadmap to future work and some open questions.« less
  2. A critical concern in data-driven decision making is to build models whose outcomes do not discriminate against some demographic groups, including gender, ethnicity, or age. To ensure non-discrimination in learning tasks, knowledge of the sensitive attributes is essential, while, in practice, these attributes may not be available due to legal and ethical requirements. To address this challenge, this paper studies a model that protects the privacy of the individuals’ sensitive information while also allowing it to learn non-discriminatory predictors. The method relies on the notion of differential privacy and the use of Lagrangian duality to design neural networks that canmore »accommodate fairness constraints while guaranteeing the privacy of sensitive attributes. The paper analyses the tension between accuracy, privacy, and fairness and the experimental evaluation illustrates the benefits of the proposed model on several prediction tasks.« less
  3. Data too sensitive to be "open" for analysis and re-purposing typically remains "closed" as proprietary information. This dichotomy undermines efforts to make algorithmic systems more fair, transparent, and accountable. Access to proprietary data in particular is needed by government agencies to enforce policy, researchers to evaluate methods, and the public to hold agencies accountable; all of these needs must be met while preserving individual privacy and firm competitiveness. In this paper, we describe an integrated legal-technical approach provided by a third-party public-private data trust designed to balance these competing interests. Basic membership allows firms and agencies to enable low-risk accessmore »to data for compliance reporting and core methods research, while modular data sharing agreements support a wide array of projects and use cases. Unless specifically stated otherwise in an agreement, all data access is initially provided to end users through customized synthetic datasets that offer a) strong privacy guarantees, b) removal of signals that could expose competitive advantage, and c) removal of biases that could reinforce discriminatory policies, all while maintaining fidelity to the original data. We find that using synthetic data in conjunction with strong legal protections over raw data strikes a balance between transparency, proprietorship, privacy, and research objectives. This legal-technical framework can form the basis for data trusts in a variety of contexts.« less
  4. Reddy, S. ; Winter, J.S. ; Padmanabhan, S. (Ed.)
    AI applications are poised to transform health care, revolutionizing benefits for individuals, communities, and health-care systems. As the articles in this special issue aptly illustrate, AI innovations in healthcare are maturing from early success in medical imaging and robotic process automation, promising a broad range of new applications. This is evidenced by the rapid deployment of AI to address critical challenges related to the COVID-19 pandemic, including disease diagnosis and monitoring, drug discovery, and vaccine development. At the heart of these innovations is the health data required for deep learning applications. Rapid accumulation of data, along with improved data quality,more »data sharing, and standardization, enable development of deep learning algorithms in many healthcare applications. One of the great challenges for healthcare AI is effective governance of these data—ensuring thoughtful aggregation and appropriate access to fuel innovation and improve patient outcomes and healthcare system efficiency while protecting the privacy and security of data subjects. Yet the literature on data governance has rarely looked beyond important pragmatic issues related to privacy and security. Less consideration has been given to unexpected or undesirable outcomes of healthcare in AI, such as clinician deskilling, algorithmic bias, the “regulatory vacuum”, and lack of public engagement. Amidst growing calls for ethical governance of algorithms, Reddy et al. developed a governance model for AI in healthcare delivery, focusing on principles of fairness, accountability, and transparency (FAT), and trustworthiness, and calling for wider discussion. Winter and Davidson emphasize the need to identify underlying values of healthcare data and use, noting the many competing interests and goals for use of health data—such as healthcare system efficiency and reform, patient and community health, intellectual property development, and monetization. Beyond the important considerations of privacy and security, governance must consider who will benefit from healthcare AI, and who will not. Whose values drive health AI innovation and use? How can we ensure that innovations are not limited to the wealthiest individuals or nations? As large technology companies begin to partner with health care systems, and as personally generated health data (PGHD) (e.g., fitness trackers, continuous glucose monitors, health information searches on the Internet) proliferate, who has oversight of these complex technical systems, which are essentially a black box? To tackle these complex and important issues, it is important to acknowledge that we have entered a new technical, organizational, and policy environment due to linked data, big data analytics, and AI. Data governance is no longer the responsibility of a single organization. Rather, multiple networked entities play a role and responsibilities may be blurred. This also raises many concerns related to data localization and jurisdiction—who is responsible for data governance? In this emerging environment, data may no longer be effectively governed through traditional policy models or instruments.« less
  5. This paper surveys recent work in the intersection of differential privacy (DP) and fairness. It reviews the conditions under which privacy and fairness may have aligned or contrasting goals, analyzes how and why DP may exacerbate bias and unfairness in decision problems and learning tasks, and describes available mitigation measures for the fairness issues arising in DP systems. The survey provides a unified understanding of the main challenges and potential risks arising when deploying privacy-preserving machine-learning or decisions-making tasks under a fairness lens.