Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher.
Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?
Some links on this page may take you to non-federal websites. Their policies may differ from this site.
-
Algorithmic recourse, or providing recommendations to individuals who receive an unfavorable outcome from an algorithmic system on how they can take action and change that outcome, is an important tool for giving individuals agency against algorithmic decision systems. Unfortunately, research on algorithmic recourse faces a fundamental challenge: there are no publicly available datasets on algorithmic recourse. In this work, we begin to explore a solution to this challenge by creating an agent-based simulation called The Game of Recourse (an homage to Conway's Game of Life) to synthesize realistic algorithmic recourse data. We designed The Game of Recourse with a focus on reliability and fairness, two areas of critical importance in socio-technical systems.more » « lessFree, publicly-accessible full text available June 9, 2025
-
In this demonstration, we present a comprehensive software library for model auditing and responsible model selection, called Virny, along with an interactive tool called VirnyView. Our library is modular and extensible, it implements a rich set of performance and fairness metrics, including novel metrics that quantify and compare model stability and uncertainty, and enables performance analysis based on multiple sensitive attributes, and their intersections. The Virny library and the VirnyView tool are available at https://github.com/DataResponsibly/Virny and https://r-ai.co/VirnyView.more » « lessFree, publicly-accessible full text available June 9, 2025
-
Database queries are often used to select and rank items as decision support for many applications. As automated decision-making tools become more prevalent, there is a growing recognition of the need to diversify their outcomes. In this paper, we define and study the problem of modifying the selection conditions of an ORDER BY query so that the result of the modified query closely fits some user-defined notion of diversity while simultaneously maintaining the intent of the original query. We show the hardness of this problem and propose a mixed-integer linear programming (MILP) based solution. We further present optimizations designed to enhance the scalability and applicability of the solution in real-life scenarios. We investigate the performance characteristics of our algorithm and show its efficiency and the usefulness of our optimizations.more » « lessFree, publicly-accessible full text available May 29, 2025
-
Concerns about the risks posed by artificial intelligence (AI) have resulted in growing interest in algorithmic transparency. While algorithmic transparency is well-studied, there is evidence that many organizations do not value implementing transparency. In this case study, we test a ground-up approach to ensuring better real-world algorithmic transparency by creating transparency influencers — motivated individuals within organizations who advocate for transparency. We held an interactive online workshop on algorithmic transparency and advocacy for 15 professionals from news, media, and journalism. We reflect on workshop design choices and presents insights from participant interviews. We found positive evidence for our approach: In the days following the workshop, three participants had done pro-transparency advocacy. Notably, one of them advocated for algorithmic transparency at an organization-wide AI strategy meeting. In the words of a participant: “if you are questioning whether or not you need to tell people [about AI], you need to tell people.”more » « lessFree, publicly-accessible full text available May 11, 2025
-
Differentially private (DP) mechanisms have been deployed in a variety of high-impact social settings (perhaps most notably by the U.S. Census). Since all DP mechanisms involve adding noise to results of statistical queries, they are expected to impact our ability to accurately analyze and learn from data, in effect trading off privacy with utility. Alarmingly, the impact of DP on utility can vary significantly among different sub-populations. A simple way to reduce this disparity is with stratification. First compute an independent private estimate for each group in the data set (which may be the intersection of several protected classes), then, to compute estimates of global statistics, appropriately recombine these group estimates. Our main observation is that naive stratification often yields high-accuracy estimates of population-level statistics, without the need for additional privacy budget. We support this observation theoretically and empirically. Our theoretical results center on the private mean estimation problem, while our empirical results center on extensive experiments on private data synthesis to demonstrate the effectiveness of stratification on a variety of private mechanisms. Overall, we argue that this straightforward approach provides a strong baseline against which future work on reducing utility disparities of DP mechanisms should be compared.more » « lessFree, publicly-accessible full text available March 25, 2025
-
The need for citizens to better understand the ethical and social challenges of algorithmic systems has led to a rapid proliferation of AI literacy initiatives. After reviewing the literature on AI literacy projects, we found that most educational practices in this area are based on teaching programming fundamentals, primarily to K-12 students. This leaves out citizens and those who are primarily interested in understanding the implications of automated decision- making systems, rather than in learning to code. To address these gaps, this article explores the methodological contributions of responsible AI education practices that focus first on stakeholders when designing learning experiences for different audiences and contexts. The article examines the weaknesses identified in current AI literacy projects, explains the stakeholder-first approach, and analyzes several responsible AI education case studies, to illustrate how such an approach can help overcome the aforementioned limitations. The results suggest that the stakeholder-first approach allows to address audiences beyond the usual ones in the field of AI literacy, and to incorporate new content and methodologies depending on the needs of the respective audiences, thus opening new avenues for teaching and research in the field.more » « less