## Poster session 1

Wed 19 10:00 PDT Wed 19 11:00 PDT

### #258 “I’m Covered in Blood”: Persistent Anti-Muslim Bias in Large Language Models

Abubakar Abid, James Zou, Maheen Farooqi

### #57 Monitoring AI Services for Misuse

Seyyed Ahmad Javadi, Chris Norval, Richard Cloete, Jatinder Singh

Given the surge in interest in AI, we now see the emergence of Artificial Intelligence as a Service (AIaaS). AIaaS entails service providers offering remote access to ML models and capabilities at arms-length’, through networked APIs. Such services will grow in popularity, as they enable access to state-of-the-art ML capabilities, ‘on demand’, ‘out of the box’, at low cost and without requiring training data or ML expertise.However, there is much public concern regarding AI. AIaaS raises particular considerations, given there is much potential for such services to be used to underpin and drive problematic, inappropriate, undesirable, controversial, or possibly even illegal applications. A key way forward is through service providers monitoring their AI services to identify potential situations of problematic use. Towards this, we elaborate the potential for ‘misuse indicators’ as a mechanism for uncovering patterns of usage behaviour warranting consideration or further investigation. We introduce a taxonomy for describing these indicators and their contextual considerations, and use exemplars to demonstrate the feasibility analysing AIaaS usage to highlight situations of possible concern. We also seek to draw more attention to AI services and the issues they raise, given AIaaS’ increasing prominence, and the general calls for the more responsible and accountable use of AI.

### #279 Accounting for Model Uncertainty in Algorithmic Discrimination

Junaid Ali, Preethi Lahoti, Krishna P. Gummadi

Traditional approaches to ensure group fairness in algorithmic decision making aim to equalize “total” error rates for different subgroups in the population. In contrast, we argue that the fairness approaches should instead focus only on equalizing errors arising due to model uncertainty (a.k.a epistemic uncertainty), caused due to lack of knowledge about the best model or due to lack of data. In other words, our proposal calls for ignoring the errors that occur due to uncertainty inherent in the data, i.e., aleatoric uncertainty. We draw a connection between predictive multiplicity and model uncertainty and argue that the techniques from predictive multiplicity could be used to identify errors made due to model uncertainty. We propose scalable convex proxies to come up with classifiers that exhibit predictive multiplicity and empirically show that our methods are comparable in performance and up to four orders of magnitude faster than the current state-of-the-art. We further pro- pose methods to achieve our goal of equalizing group error rates arising due to model uncertainty in algorithmic decision making and demonstrate the effectiveness of these methods using synthetic and real-world datasets

### #93 Beyond Reasonable Doubt: Improving Fairness in Budget-Constrained Decision-Making Using Confidence Thresholds

Michiel Bakker, Duy Patrick Tu, Krishna Gummadi, Alex Pentland, Kush Varshney, Adrian Weller

Prior work on fairness in machine learning has focused on settings where all the information needed about each individual is readily available. However, in many applications, further information may be acquired at a cost. For example, when assessing a customer’s creditworthiness, a bank initially has access to a limited set of information but progressively improves the assessment by acquiring additional information before making a final decision. In such settings, we posit that a fair decision maker may want to ensure that decisions for all individuals are made with similar expected error rate, even if the features acquired for the individuals are different. We show that a set of carefully chosen confidence thresholds can not only effectively redistribute an information budget according to each individual’s needs, but also serve to address individual and group fairness concerns simultaneously. Finally, using two public datasets, we confirm the effectiveness of our methods and investigate the limitations.

### #160 Ensuring Fairness under Prior Probability Shifts

Arpita Biswas, Suvam Mukherjee

Prior probability shift is a phenomenon where the training and test datasets differ structurally within population subgroups. This phenomenon can be observed in the yearly records of several real-world datasets, for example, recidivism records and medical expenditure surveys. If unaccounted for, such shifts can cause the predictions of a classifier to become unfair towards specific population subgroups. While the fairness notion called Proportional Equality (PE) accounts for such shifts, a procedure to ensure PE-fairness was unknown. In this work, we design an algorithm, called CAPE, that ensures fair classification under such shifts. We introduce a metric, called prevalence difference, which CAPE attempts to minimize in order to achieve fairness under prior probability shifts. We theoretically establish that this metric exhibits several properties that are desirable for a fair classifier. We evaluate the efficacy of CAPE via a thorough empirical evaluation on synthetic datasets. We also compare the performance of CAPE with several state-of-the-art fair classifiers on real-world datasets like COMPAS (criminal risk assessment) and MEPS (medical expenditure panel survey). The results indicate that CAPE ensures a high degree of PE-fairness in its predictions, while performing well on other important metrics.

### #220 Envisioning Communities: A Participatory Approach towards AI for Social Good

Elizabeth Bondi, Lily Xu, Diana Acosta-Navas, Jackson A. Killian

Research in artificial intelligence (AI) for social good presupposes some definition of social good, but potential definitions have been seldom suggested and never agreed upon. The normative question of what AI for social good research should be "for" is not thoughtfully elaborated, or is frequently addressed with a utilitarian outlook that prioritizes the needs of the majority over those who have been historically marginalized, brushing aside realities of injustice and inequity. We argue that AI for social good ought to be assessed by the communities that the AI system will impact, using as a guide the capabilities approach, a framework to measure the ability of different policies to improve human welfare equity. Furthermore, we lay out how AI research has the potential to catalyze social progress by expanding and equalizing capabilities. We show how the capabilities approach aligns with a participatory approach for the design and implementation of AI for social good research in a framework we introduce called PACT, in which community members affected should be brought in as partners and their input prioritized throughout the project. We conclude by providing an incomplete set of guiding questions for carrying out such participatory AI research in a way that elicits and respects a community’s own definition of social good.

### #72 AI Alignment and Human Reward

Patrick Butlin.

According to a prominent approach to AI alignment, AI agents should be built to learn and promote human values. However, humans value things in several different ways: we have desires and preferences of various kinds, and if we engage in reinforcement learning, we also have reward functions. One research project to which this approach gives rise is therefore to say which of these various classes of human values should be promoted. This paper takes on part of this project by assessing the proposal that human reward functions should be the target for AI alignment. There is some reason to believe that powerful AI agents which were aligned to values of this form would help us to lead good lives, but there is also considerable uncertainty about this claim, arising from unresolved empirical and conceptual issues in human psychology.

### #83 Modeling and Guiding the Creation of Ethical Human-AI Teams

Christopher Flathmann, Beau Schelble, Rui Zhang, Nathan McNeese

With artificial intelligence continuing to advance, so too do the ethical concerns that can potentially negatively impact humans and the greater society. When these systems begin to interact with humans, these concerns become much more complex and much more important. The field of human-AI teaming provides a relevant example of how AI ethics can have significant and continued effects on humans. This paper reviews research in ethical artificial intelligence, as well as ethical teamwork through the lens of the rapidly advancing field of human-AI teaming, resulting in a model demonstrating the requirements and outcomes of building ethical human-AI teams. The model is created to guide the prioritization of ethics in human-AI teaming by outlining the ethical teaming process, outcomes of ethical teams, and external requirements necessary to ensure ethical human-AI teams. A final discussion is presented on how the developed model will influence the implementation of AI teammates, as well as the development of policy and regulation surrounding the domain in the coming years.

### #252 What’s Fair about Individual Fairness?

Will Fleisher

One of the main lines of research in algorithmic fairness involves individual fairness (IF) methods. Individual fairness is motivated by an intuitive principle, similar treatment, which requires that similar individuals be treated similarly. IF offers a precise account of this principle using distance metrics to evaluate the similarity of individuals. Proponents of individual fairness have argued that it gives the correct definition of algorithmic fairness, and that it should therefore be preferred to other methods for determining fairness. I argue that individual fairness cannot serve as a definition of fairness. Moreover, IF methods should not be given priority over other fairness methods, nor used in isolation from them. To support these conclusions, I describe four in-principle problems for individual fairness as a definition and as a method for ensuring fairness: (1) counterexamples show that similar treatment (and therefore IF) are insufficient to guarantee fairness; (2) IF methods for learning similarity metrics are at risk of encoding human implicit bias; (3) IF requires prior moral judgments, limiting its usefulness as a guide for fairness and undermining its claim to define fairness; and (4) the incommensurability of relevant moral values makes similarity metrics impossible for many tasks. In light of these limitations, I suggest that individual fairness cannot be a definition of fairness, and instead should be seen as one tool among several for ameliorating algorithmic bias.

### #36 Learning to Generate Fair Clusters from Demonstrations

Sainyam Galhotra, Sandhya Saisubramanian, Shlomo Zilberstein

Fair clustering is the process of grouping similar entities together, while satisfying a mathematically well-defined fairness metric as a constraint. Due to the practical challenges in precise model specification, the prescribed fairness constraints are often incomplete and act as proxies to the intended fairness requirement. Clustering with proxies may lead to biased outcomes when the system is deployed. We examine how to identify the intended fairness constraint for a problem based on limited demonstrations from an expert. Each demonstration is a clustering over a subset of the data. We present an algorithm to identify the fairness metric from demonstrations and generate clusters using existing off-the-shelf clustering techniques, and analyze its theoretical properties. To extend our approach to novel fairness metrics for which clustering algorithms do not currently exist, we present a greedy method for clustering. Additionally, we investigate how to generate interpretable solutions using our approach. Empirical evaluation on three real-world datasets demonstrates the effectiveness of our approach in quickly identifying the underlying fairness and interpretability constraints, which are then used to generate fair and interpretable clusters.

### #31 Ethical Obligations to Provide Novelty

Paige Golden, David Danks

TikTok is a popular platform that enables users to see tailored content feeds, particularly short videos with novel content. In recent years, TikTok has been criticized at times for presenting users with overly homogenous feeds, thereby reducing the diversity of content with which each user engages. In this paper, we consider whether TikTok has an ethical obligation to employ a novelty bias in its content recommendation engine. We explicate the principal morally relevant values and interests of key stakeholders, and observe that key empirical questions must be answered before a precise recommendation can be provided. We argue that TikTok’s own values and interests mean that its actions should be largely driven by the values and interests of its users and creators. Unlike some other content platforms, TikTok’s ethical obligations are not at odds with the values of its users, and so whether it is obligated to include a novelty bias depends on what will actually advance its users’ interests.

### #2 An AI Ethics Course Highlighting Explicit Ethical Agents

Nancy Green

This is an experience report describing a pilot AI Ethics course for undergraduate computer science majors. In addition to teaching students about different ethical approaches and using them to analyze ethical issues, the course covered how ethics has been incorporated into the implementation of explicit ethical agents, and required students to implement an explicit ethical agent for a simple application. This report describes the course objectives and design, the topics covered, and a qualitative evaluation with suggestions for future offerings of the courses.

### #147 The Dangers of Drowsiness Detection: Differential Performance, Downstream Impact, and Misuses

Jakub Grzelak, Martim Brandao

Drowsiness and fatigue are important factors in driving safety and work performance. This has motivated academic research into detecting drowsiness, and sparked interest in the deployment of related products in the insurance and work-productivity sectors. In this paper we elaborate on the potential dangers of using such algorithms. We first report on an audit of performance bias across subject gender and ethnicity, identifying which groups would be disparately harmed by the deployment of a state-of-the-art drowsiness detection algorithm. We discuss some of the sources of the bias, such as the lack of robustness of facial analysis algorithms to face occlusions, facial hair, or skin tone. We then identify potential downstream harms of this performance bias, as well as potential misuses of drowsiness detection technology—focusing on driving safety and experience, insurance cream-skimming and coverage-avoidance, worker surveillance, and job precarity.

### #48 Situated Accountability: Ethical Principles, Certification Standards, and Explanation Methods in Applied AI

Anne Henriksen, Simon Enni, Anja Bechmann

Artificial intelligence (AI) has the potential to benefit humans and society by its employment in important sectors. However, the risks of negative consequences have underscored the importance of accountability for AI systems, their outcomes, and the users of such systems. In recent years, various accountability mechanisms have been put forward in pursuit of the responsible design, development, and use of AI. In this article, we provide an in-depth study of three such mechanisms, as we analyze Scandinavian AI developers’ encounter with (1) ethical principles, (2) certification standards, and (3) explanation methods. By doing so, we contribute to closing a gap in the literature between discussions of accountability on the research and policy level, and accountability as a responsibility put on the shoulders of developers in practice. Our study illustrates important flaws in the current enactment of accountability as an ethical and social value which, if left unchecked, risks undermining the pursuit of responsible AI. By bringing attention to these flaws, the article signals where further work is needed in order to build effective accountability systems for AI.

### #190 Exciting, Useful, Worrying, Futuristic: Public Perception of Artificial Intelligence in 8 Countries

Patrick Gage Kelley, Yongwei Yang, Courtney Heldreth, Christopher Moessner, Aaron Sedley, Andreas Kramm, David T. Newman, Allison Woodruff

As the influence and use of artificial intelligence (AI) have grown and its transformative potential has become more apparent, many questions have been raised regarding the economic, political, social, and ethical implications of its use. Public opinion plays an important role in these discussions, influencing product adoption, commercial development, research funding, and regulation. In this paper we present results of an in-depth survey of public opinion of artificial intelligence conducted with 10,005 respondents spanning eight countries and six continents. We report widespread perception that AI will have significant impact on society, accompanied by strong support for the responsible development and use of AI, and also characterize the public’s sentiment towards AI with four key themes (exciting, useful, worrying, and futuristic) whose prevalence distinguishes response to AI in different countries.

### #212 Age Bias in Emotion Detection: Analysis of Facial Emotion Recognition Performance on Varying Age Groups

Eugenia Kim, De’Aira Bryant, Deepak Srikanth, Ayanna Howard

The growing potential for facial emotion recognition (FER) technology has encouraged expedited development at the cost of rigorous validation. Many of its use-cases may also impact the diverse global community as FER becomes embedded into domains ranging from education to security to healthcare. Yet, prior work has highlighted that FER can exhibit both gender and racial biases like other facial analysis techniques. As a result, bias-mitigation research efforts have mainly focused on tackling gender and racial disparities, while other demographic related biases, such as age, have seen less progress. This work seeks to examine the performance of state of the art commercial FER technology on expressive images of men and women from three distinct age groups. We utilize four different commercial FER systems in a black box methodology to evaluate how six emotions – anger, disgust, fear, happiness, neutrality, and sadness – are correctly detected by age group. We further investigate how algorithmic changes over the last year have affected system performance. Our results found that all four commercial FER systems most accurately perceived emotion in images of young adults and least accurately in images of older adults. This trend was observed for analyses conducted in 2019 and 2020. However, little to no gender disparities were observed in either year. While older adults may not have been the initial target consumer of FER technology, statistics show the demographic is quickly growing more keen to applications that use such systems. Our results demonstrate the importance of considering various demographic subgroups during FER system validation and the need for inclusive, intersectional algorithmic developmental practices.

### #172 Becoming Good at AI for Good

Meghana Kshirsagar, Caleb Robinson, Siyu Yang, Shahrzad Gholami, Ivan Klyuzhin, Sumit Mukherjee, Md Nasir, Anthony Ortiz, Felipe Oviedo Perhavec, Darren Tanner, Anusua Trivedi, Yixi Xu, Ming Zhong, Bistra Dilkina, Rahul Dodhia, Juan Lavista Ferres

AI for good (AI4G) projects involve developing and applying artificial intelligence (AI) based solutions to further goals in areas such as sustainability, health, humanitarian aid, and social justice. Developing and deploying such solutions must be done in collaboration with partners who are experts in the domain in question and who already have experience in making progress towards such goals. Based on our experiences, we detail the different aspects of this type of collaboration broken down into four high-level categories: communication, data, modeling, and impact, and distill eleven takeaways to guide such projects in the future. We briefly describe two case studies to illustrate how some of these takeaways were applied in practice during our past collaborations.

### #245 RAWLSNET: Altering Bayesian Networks to Encode Rawlsian Fair Equality of Opportunity

David Liu, Zohair Shafi, William Fleisher, Tina Eliassi-Rad, Scott Alfeld

We present RAWLSNET, a system for altering Bayesian Network (BN) models to satisfy the Rawlsian principle of fair equality of opportunity (FEO). RAWLSNET’s BN models generate aspirational data distributions: data generated to reflect an ideally fair, FEO-satisfying society. FEO states that everyone with the same talent and willingness to use it should have the same chance of achieving advantageous social positions (e.g., employment), regardless of their background circumstances (e.g., socioeconomic status). Satisfying FEO requires alterations to social structures such as school assignments. Our paper describes RAWLSNET, a method which takes as input a BN representation of an FEO application and alters the BN’s parameters so as to satisfy FEO when possible, and minimize deviation from FEO otherwise. We also offer guidance for applying RAWLSNET, including on recognizing proper applications of FEO. We demonstrate the use of RAWLSNET with publicly available data sets. RAWLSNET’s altered BNs offer the novel capability of generating aspirational data for FEO-relevant tasks. Aspirational data are free from biases of real-world data, and thus are useful for recognizing and detecting sources of unfairness in machine learning algorithms besides biased data.

### #275 Fair Equality of Chances: Fairness for Statistical Prediction-Based Decision-Making

Michele Loi, Anders Herlitz, Heidari Hoda

This paper presents a fairness principle that can be used to evaluate decision-making based on predictions. We propose that a decision rule for decision-making based on predictions is unfair when the individuals directly subjected to the implications of the decision do not enjoy fair equality of chances. We define fair equality of chances as obtaining if and only if the individuals who are equal with respect to the features that justify inequalities in outcomes have the same statistical prospects of being benefited or harmed, irrespective of their morally irrelevant traits. The paper characterizes – in a formal way – the way in which luck can be allowed to impact outcomes without the process being unfair. This fairness principle can be used to evaluate decision-making based on predictions, a kind of decision-making that is becoming increasingly important to analyze in light of the growing prevalence of algorithmic decision-making. It can be used to evaluate decision-making rules based on different normative theories, and is compatible with a broad range of normative views according to which inequalities due to brute luck can be fair.

### #99 More Similar Values, More Trust? The Effect of Value Similarity on Trust in Human-Agent Interaction

Siddharth Mehrotra, Catholijn M. Jonker, Myrthe L. Tielman

As AI systems are increasingly involved in decision making, it also becomes important that they elicit appropriate levels of trust from their users. To achieve this, it is first important to understand which factors influence trust in AI. We identify that a research gap exists regarding the role of personal values in trust in AI. Therefore, this paper studies how human and agent Value Similarity (VS) influences a human’s trust in that agent. To explore this, 89 participants teamed up with five different agents, which were designed with varying levels of value similarity to that of the participants. In a within-subjects, scenario-based experiment, agents gave suggestions on what to do when entering the building to save a hostage. We analyzed the agent’s scores on subjective value similarity, trust and qualitative data from open-ended questions. Our results show that agents rated as having more similar values also scored higher on trust, indicating a positive effect between the two. With this result, we add to the existing understanding of human-agent trust by providing insight into the role of value-similarity.

### #134 Causal Multi-Level Fairness

Vishwali Mhasawade, Rumi Chunara

Algorithmic systems are known to impact marginalized groups severely, and more so, if all sources of bias are not considered. While work in algorithmic fairness to-date has primarily focused on addressing discrimination due to individually linked attributes, social science research elucidates how some properties we link to individuals can be conceptualized as having causes at macro (e.g. structural) levels, and it may be important to be fair to attributes at multiple levels. For example, instead of simply considering race as a causal, protected attribute of an individual, the cause may be distilled as perceived racial discrimination an individual experiences, which in turn can be affected by neighborhood-level factors. This multi-level conceptualization is relevant to questions of fairness, as it may not only be important to take into account if the individual belonged to another demographic group, but also if the individual received advantaged treatment at the macro-level. In this paper, we formalize the problem of multi-level fairness using tools from causal inference in a manner that allows one to assess and account for effects of sensitive attributes at multiple levels. We show importance of the problem by illustrating residual unfairness if macro-level sensitive attributes are not accounted for, or included without accounting for their multi-level nature. Further, in the context of a real-world task of predicting income based on macro and individual-level attributes, we demonstrate an approach for mitigating unfairness, a result of multi-level sensitive attributes.

### #43 Disparate Impact of Artificial Intelligence Bias in Ridehailing Economy’s Price Discrimination Algorithms

Akshat Pandey, Aylin Caliskan

Ridehailing applications that collect mobility data from individuals to inform smart city planning predict each trip’s fare pricing with automated algorithms that rely on artificial intelligence (AI). This type of AI algorithm, namely a price discrimination algorithm, is widely used in the industry’s black box systems for dynamic individualized pricing. Lacking transparency, studying such AI systems for fairness and disparate impact has not been possible without access to data used in generating the outcomes of price discrimination algorithms. Recently, in an effort to enhance transparency in city planning, the city of Chicago regulation mandated that transportation providers publish anonymized data on ridehailing. As a result, we present the first large-scale measurement of the disparate impact of price discrimination algorithms used by ridehailing applications. The application of random effects models from the meta-analysis literature combines the city-level effects of AI bias on fare pricing from census tract attributes, aggregated from the American Community Survey. An analysis of 100 million ridehailing samples from the city of Chicago indicates a significant disparate impact in fare pricing of neighborhoods due to AI bias learned from ridehailing utilization patterns associated with demographic attributes. Neighborhoods with larger non-white populations, higher poverty levels, younger residents, and high education levels are significantly associated with higher fare prices, with combined effect sizes, measured in Cohen’s d, of -0.32, -0.28, 0.69, and 0.24 for each demographic, respectively. Further, our methods hold promise for identifying and addressing the sources of disparate impact in AI algorithms learning from datasets that contain U.S. geolocations.

### #138 Understanding the Representation and Representativeness of Age in AI Data Sets

Joon Sung Park, Michael S. Bernstein, Robin N. Brewer, Ece Kamar, Meredith Ringel Morris

A diverse representation of different demographic groups in AI training data sets is important in ensuring that the models will work for a large range of users. To this end, recent efforts in AI fairness and inclusion have advocated for creating AI data sets that are well-balanced across race, gender, socioeconomic status, and disability status. In this paper, we contribute to this line of work by focusing on the representation of age by asking whether older adults are represented proportionally to the population at large in AI data sets. We examine publicly-available information about 92 face data sets to understand how they codify age as a case study to investigate how the subjects’ ages are recorded and whether older generations are represented. We find that older adults are very under-represented; five data sets in the study that explicitly documented the closed age intervals of their subjects included older adults (defined as older than 65 years), while only one included oldest-old adults (defined as older than 85 years). Additionally, we find that only 24 of the data sets include any age-related information in their documentation or metadata, and that there is no consistent method followed across these data sets to collect and record the subjects’ ages. We recognize the unique difficulties in creating representative data sets in terms of age, but raise it as an important dimension that researchers and engineers interested in inclusive AI should consider.

### #268 Face Mis-ID: Interrogating Facial Recognition Harms with Community Organizers Using an Interactive Demo

Daniella Raz, Corinne Bintz, Vivian Guetler, Michael Katell, Dharma Dailey, Bernease Herman, P. M. Krafft, Meg Young

This paper reports on the making of an interactive demo to illustrate algorithmic bias in facial recognition. Facial recognition technology has been demonstrated to be more likely to misidentify women and minoritized people. This risk, among others, has elevated facial recognition into policy discussions across the country, where many jurisdictions have already passed bans on its use. Whereas scholarship on the disparate impacts of algorithmic systems is growing, general public awareness of this set of problems is limited in part by the illegibility of machine learning systems to non-specialists. Inspired by discussions with community organizers advocating for tech fairness issues, we created the Face Mis-ID Demo to reveal the algorithmic functions behind facial recognition technology and to demonstrate its risks to policymakers and members of the community. In this paper, we share the design process behind this interactive demo, its form and function, and the design decisions that honed its accessibility, toward its use for improving legibility of algorithmic systems and awareness of the sources of their disparate impacts.

### #233 GAEA: Graph Augmentation for Equitable Access via Reinforcement Learning

Govardana Sachithanandam Ramachandran, Ivan Brugere, Lav R. Varshney, Caiming Xiong

Disparate access to resources by different subpopulations is a prevalent issue in societal and sociotechnical networks. For example, urban infrastructure networks may enable certain racial groups to more easily access resources such as high-quality schools, grocery stores, and polling places. Similarly, social networks within universities and organizations may enable certain groups to more easily access people with valuable information or influence. Here we introduce a new class of problems, Graph Augmentation for Equitable Access (GAEA), to enhance equity in networked systems by editing graph edges under budget constraints. We prove such problems are NP-hard, and cannot be approximated within a factor of $(1-\tfrac{1}{3e})$. We develop a principled, sample- and time- efficient Markov Reward Process (MRP)-based mechanism design framework for GAEA. Our algorithm outperforms baselines on a diverse set of synthetic graphs. We further demonstrate the method on real-world networks, by merging public census, school, and transportation datasets for the city of Chicago and applying our algorithm to find human-interpretable edits to the bus network that enhance equitable access to high-quality schools across racial groups. Further experiments on Facebook networks of universities yield sets of new social connections that would increase equitable access to certain attributed nodes across gender groups.

### #184 Does Fair Ranking Improve Minority Outcomes? Understanding the Interplay of Human and Algorithmic Biases in Online Hiring

Tom Sühr, Sophie Hilgard, Himabindu Lakkaraju

Ranking algorithms are being widely employed in various online hiring platforms including LinkedIn, TaskRabbit, and Fiverr.Prior research has demonstrated that ranking algorithms employed by these platforms are prone to a variety of undesirable biases, leading to the proposal of fair ranking algorithms (e.g., Det-Greedy) which increase exposure of underrepresented candidates. However, there is little to no work that explores whether fair ranking algorithms actually improve real world outcomes (e.g., hiring decisions) for underrepresented groups. Furthermore, there is no clear understanding as to how other factors (e.g., job context, inherent biases of the employers) may impact the efficacy of fair ranking in practice.In this work, we analyze various sources of gender biases in online hiring platforms, including the job context and inherent biases of employers and establish how these factors interact with ranking algorithms to affect hiring decisions. To the best of our knowledge, this work makes the first attempt at studying the interplay between the aforementioned factors in the context of online hiring. We carry out a large-scale user study simulating online hiring scenarios with data from TaskRabbit, a popular online freelancing site. Our results demonstrate that while fair ranking algorithms generally improve the selection rates of underrepresented minorities, their effectiveness relies heavily on the job contexts and candidate profiles.

### #111 Governing Algorithmic Systems with Impact Assessments: Six Observations

Elizabeth Anne Watkins, Emanuel Moss, Jacob Metcalf, Ranjit Singh, Madeleine Elish

Algorithmic decision-making and decision-support systems (ADS) are gaining influence over how society distributes resources, administers justice, and provides access to opportunities. Yet collectively we do not adequately study how these systems affect people or document the actual or potential harms resulting from their integration with important social functions. This is a significant challenge for computational justice efforts of measuring and governing AI systems. Impact assessments are often used as instruments to create accountability relationships and grant some measure of agency and voice to communities affected by projects with environmental, financial, and human rights ramifications. Applying these tools—through Algorithmic Impact Assessments (AIA)—is a plausible way to establish accountability relationships for ADSs. At the same time, what an AIA would entail remains under-specified; they raise as many questions as they answer. Choices about the methods, scope, and purpose of AIAs structure the conditions of possibility for AI governance. In this paper, we present our research on the history of impact assessments across diverse domains, through a sociotechnical lens, to present six observations on how they co-constitute accountability. Decisions about what type of effects count as an impact; when impacts are assessed; whose interests are considered; who is invited to participate; who conducts the assessment; how assessments are made publicly available, and what the outputs of the assessment might be; all shape the forms of accountability that AIAs engender. Because AlAs are still an incipient governance strategy, approaching them as social constructions that do not require a single or universal approach offers a chance to produce interventions that emerge from careful deliberation.

### #119 A Human-in-the-Loop Framework to Construct Context-Aware Mathematical Notions of Outcome Fairness

Mohammad Yaghini, Hoda Heidari, Andreas Krause

Existing mathematical notions of fairness fail to account for the \textbf{context} of decision-making. We argue that moral consideration of contextual factors is an inherently \emph{human} task. So we present a framework to learn \emph{context-aware} mathematical formulations of fairness by eliciting people’s \emph{situated fairness assessments}. Our family of fairness notions corresponds to a new interpretation of economic models of \emph{Equality of Opportunity (EOP)}, and it includes most existing notions of fairness as special cases. Our \emph{human-in-the-loop} approach is designed to learn the appropriate parameters of the EOP family by utilizing human responses to pair-wise questions about decision subjects’ \emph{circumstance} and \emph{deservingness}, and the \emph{harm/benefit} imposed on them. We illustrate our framework in a hypothetical criminal risk assessment scenario by conducting a series of human-subject experiments on Amazon Mechanical Turk. Our work takes an important initial step toward empowering stakeholders to have a voice in the formulation of fairness for Machine Learning.