Search

In this paper, we consider the problem of batch classification and propose a novel framework for achieving fairness in such settings. The problem of batch classification involves selection of a set of individuals, often encountered in real-world scenarios such as job recruitment, college admissions etc. This is in contrast to a typical classification problem, where each candidate in the test set is considered separately and independently. In such scenarios, achieving the same acceptance rate (i.e., probability of the classifier assigning positive class) for each group (membership determined by the value of sensitive attributes such as gender, race etc.) is often not desirable, and the regulatory body specifies a different acceptance rate for each group. The existing fairness enhancing methods do not allow for such specifications and hence are unsuited for such scenarios. In this paper, we define a configuration model whereby the acceptance rate of each group can be regulated and further introduce a novel batch-wise fairness post-processing framework using the classifier confidence-scores. We deploy our framework across four real-world datasets and two popular notions of fairness, namely demographic parity and equalized odds. In addition to consistent performance improvements over the competing baselines, the proposed framework allows flexibility and significant speed-up. It can also seamlessly incorporate multiple overlapping sensitive attributes. To further demonstrate the generalizability of our framework, we deploy it to the problem of fair gerrymandering where it achieves a better fairness-accuracy trade-off than the existing baseline method. [ABSTRACT FROM AUTHOR]

Fairness has become a critical value online, and the latest studies consider it in many problems. In recommender systems, fairness is important since the visibility of items is controlled by systems. Previous fairness-aware recommender systems assume that sufficient relationship data between users and items are available. However, it is common that new users and items are frequently introduced, and they have no relationship data yet. In this paper, we study recommendation methods to enhance fairness in a cold-start state. Fairness is more significant when the preference of a user or the popularity of an item is unknown. We propose a meta-learning-based cold-start recommendation framework called FaRM to alleviate the unfairness of recommendations. The proposed framework consists of three steps. We first propose a fairness-aware meta-path generation method to eliminate bias in sensitive attributes. In addition, we construct fairness-aware user representations through the meta-path aggregation approach. Then, we propose a novel fairness objective function and introduce a joint learning method to minimize the trade-off between relevancy and fairness. In extensive experiments with various cold-start scenarios, it is shown that FaRM is significantly superior in fairness performance while preserving relevance accuracy over previous work. [ABSTRACT FROM AUTHOR]

Due to the pressure of the ecological environment and government, it is incumbent for enterprises to undertake corporate social responsibility (CSR). However, during the recycling process, awareness of equity concerns due to the distribution of benefits among members has intensified, and it is crucial to resolve channel conflicts and design a reasonable cooperation model to recycle used power batteries. Therefore, this paper constructs a closed-loop supply chain composed of power battery manufacturer, retailer and third-party collector based on the consideration of cascade utilization, in order to study the impact of the level of CSR and the degree of fairness concern on the decision-making of channel participants. Our research shows that: (1) Fairness concern behavior adversely affects the supply chain, which raises the sales price and reduces the collecting rate and the utility of the supply chain as a whole. (2) Undertaking CSR is beneficial to the development of the power battery market, and also helps to reduce the sense of unfairness among third-party recycling companies. (3) The cost-sharing contract effectively coordinates the distribution of supply chain benefits and improves the recycling rate. Finally, we further verify the correctness of the conclusions through numerical studies. [ABSTRACT FROM AUTHOR]

In this paper, we show that fairness can evolve in the divide-a-lottery game which is more general than the divide-a-dollar game by using an indirect evolutionary approach. In the divide-a-lottery game, the size of a pie is uncertain. Two players sequentially bid for a share and they get their bid if the allocation based on the bids turns out to be feasible and otherwise neither gets anything. In this game, rational players over-compete for a higher share, resulting in a high probability of failure in agreement, whereas fair players who dislike the disparity between shares lower their bids thereby reducing the failure probability and thus increasing the expected payoff. As a result, fairness strictly dominates rationality. This is the mechanism through which fairness evolves. However, this result is not robust against even a slight uncertainty about the opponent's type. Surprisingly, we show a contrasted simulation result that only rational players who are strictly dominated by fair players survive evolutionarily for most of the parameter values if players have even a slight chance of not knowing the opponent's type. Our simulation results in a local interaction model in which players only know the type of closer neighbors capture both insights and demonstrate that moderate proportions of both types coexist evolutionarily over time, and that the population average fitness of this polymorphic population is higher than monomorphic population consisting only of fair types or rational types. [ABSTRACT FROM AUTHOR]

The use of data-driven decision support by public agencies is becoming more widespread and already influences the allocation of public resources. This raises ethical concerns, as it has adversely affected minorities and historically discriminated groups. In this paper, we use an approach that combines statistics and data-driven approaches with dynamical modeling to assess long-term fairness effects of labor market interventions. Specifically, we develop and use a model to investigate the impact of decisions caused by a public employment authority that selectively supports job-seekers through targeted help. The selection of who receives what help is based on a data-driven intervention model that estimates an individual's chances of finding a job in a timely manner and rests upon data that describes a population in which skills relevant to the labor market are unevenly distributed between two groups (e.g., males and females). The intervention model has incomplete access to the individual's actual skills and can augment this with knowledge of the individual's group affiliation, thus using a protected attribute to increase predictive accuracy. We assess this intervention model's dynamics—especially fairness-related issues and trade-offs between different fairness goals- over time and compare it to an intervention model that does not use group affiliation as a predictive feature. We conclude that in order to quantify the trade-off correctly and to assess the long-term fairness effects of such a system in the real-world, careful modeling of the surrounding labor market is indispensable. [ABSTRACT FROM AUTHOR]

In heterogeneous wireless networks, random packet loss and high latency lead to conventional TCP variants performing unsatisfactorily in the case of competing communications. Especially on high-latency wireless links, conventional TCP variants are unable to estimate congestion degrees accurately for fine-grained congestion control because of the effects of random packet loss and delay oscillations. This paper proposes a TCP variant at the sender side to identify congestion degrees, namely TCP-WBQ, which quickly responses to the real congestion and effectively shields against random packet loss and oscillations of latency time. The proposed algorithm of congestion control firstly constructs a backlog-queue model based on the dynamics of the congestion window, and deduces the two bounds of the model which delimit oscillations of the backlog queue for non-congestion and random packet loss respectively. TCP-WBQ detects congestion degrees more accurately and thus implements the corresponding schemes of adjusting the congestion window, maintaining a tradeoff between high throughputs and congestion avoidance. The comprehensive simulations show that TCP-WBQ works efficiently in bandwidth utilization with single and multiple bottleneck scenarios, and achieves high performance and competitive fairness in heterogeneous wireless networks. [ABSTRACT FROM AUTHOR]

Abstract Fairness has become a critical value online, and the latest studies consider it in many problems. In recommender systems, fairness is important since the visibility of items is controlled by systems. Previous fairness-aware recommender systems assume that sufficient relationship data between users and items are available. However, it is common that new users and items are frequently introduced, and they have no relationship data yet. In this paper, we study recommendation methods to enhance fairness in a cold-start state. Fairness is more significant when the preference of a user or the popularity of an item is unknown. We propose a meta-learning-based cold-start recommendation framework called FaRM to alleviate the unfairness of recommendations. The proposed framework consists of three steps. We first propose a fairness-aware meta-path generation method to eliminate bias in sensitive attributes. In addition, we construct fairness-aware user representations through the meta-path aggregation approach. Then, we propose a novel fairness objective function and introduce a joint learning method to minimize the trade-off between relevancy and fairness. In extensive experiments with various cold-start scenarios, it is shown that FaRM is significantly superior in fairness performance while preserving relevance accuracy over previous work.

Governance quality refers to how well the processes and institutions of public governance function and is widely recognized as having an important influence on human well-being. We developed and tested a theoretical model that elucidates the relationship between governance quality and the subjective well-being of residents in China by revealing the serial mediation effects of perceived social fairness and trust in government. Using data from the nationally representative Chinese Social Survey conducted in 2021 (n = 5019), we performed structural equation modeling to empirically examine our hypotheses. The results indicated that governance quality exerted a significant positive fully indirect impact on subjective well-being through perceived social fairness, trust in government, and their serial mediation effects. This study contributes to the literature by providing valuable insights into the determinants of subjective well-being and highlighting the serial mediating roles of perceived social fairness and trust in government in the relationship between governance quality and subjective well-being. The findings also provide practical insights for policymakers, as they indicate that promoting perceived social fairness and fostering trust in government are essential to translate governance quality into subjective well-being. [ABSTRACT FROM AUTHOR]

A significant level of stigma and inequality exists in mental healthcare, especially in under-served populations. Inequalities are reflected in the data collected for scientific purposes. When not properly accounted for, machine learning (ML) models learned from data can reinforce these structural inequalities or biases. Here, we present a systematic study of bias in ML models designed to predict depression in four different case studies covering different countries and populations. We find that standard ML approaches regularly present biased behaviors. We also show that mitigation techniques, both standard and our own post-hoc method, can be effective in reducing the level of unfair bias. There is no one best ML model for depression prediction that provides equality of outcomes. This emphasizes the importance of analyzing fairness during model selection and transparent reporting about the impact of debiasing interventions. Finally, we also identify positive habits and open challenges that practitioners could follow to enhance fairness in their models. [ABSTRACT FROM AUTHOR]

In recent years, the problem of addressing fairness in machine learning (ML) and automatic decision making has attracted a lot of attention in the scientific communities dealing with artificial intelligence. A plethora of different definitions of fairness in ML have been proposed, that consider different notions of what is a "fair decision" in situations impacting individuals in the population. The precise differences, implications and "orthogonality" between these notions have not yet been fully analyzed in the literature. In this work, we try to make some order out of this zoo of definitions. [ABSTRACT FROM AUTHOR]

Prediction models are commonly used to estimate risk for cardiovascular diseases, to inform diagnosis and management. However, performance may vary substantially across relevant subgroups of the population. Here we investigated heterogeneity of accuracy and fairness metrics across a variety of subgroups for risk prediction of two common diseases: atrial fibrillation (AF) and atherosclerotic cardiovascular disease (ASCVD). We calculated the Cohorts for Heart and Aging in Genomic Epidemiology Atrial Fibrillation (CHARGE-AF) score for AF and the Pooled Cohort Equations (PCE) score for ASCVD in three large datasets: Explorys Life Sciences Dataset (Explorys, n = 21,809,334), Mass General Brigham (MGB, n = 520,868), and the UK Biobank (UKBB, n = 502,521). Our results demonstrate important performance heterogeneity across subpopulations defined by age, sex, and presence of preexisting disease, with fairly consistent patterns across both scores. For example, using CHARGE-AF, discrimination declined with increasing age, with a concordance index of 0.72 [95% CI 0.72–0.73] for the youngest (45–54 years) subgroup to 0.57 [0.56–0.58] for the oldest (85–90 years) subgroup in Explorys. Even though sex is not included in CHARGE-AF, the statistical parity difference (i.e., likelihood of being classified as high risk) was considerable between males and females within the 65–74 years subgroup with a value of − 0.33 [95% CI − 0.33 to − 0.33]. We also observed weak discrimination (i.e., < 0.7) and suboptimal calibration (i.e., calibration slope outside of 0.7–1.3) in large subsets of the population; for example, all individuals aged 75 years or older in Explorys (17.4%). Our findings highlight the need to characterize and quantify the behavior of clinical risk models within specific subpopulations so they can be used appropriately to facilitate more accurate, consistent, and equitable assessment of disease risk. [ABSTRACT FROM AUTHOR]

The recent release of large-scale healthcare datasets has greatly propelled the research of data-driven deep learning models for healthcare applications. However, due to the nature of such deep black-boxed models, concerns about interpretability, fairness, and biases in healthcare scenarios where human lives are at stake call for a careful and thorough examination of both datasets and models. In this work, we focus on MIMIC-IV (Medical Information Mart for Intensive Care, version IV), the largest publicly available healthcare dataset, and conduct comprehensive analyses of interpretability as well as dataset representation bias and prediction fairness of deep learning models for in-hospital mortality prediction. First, we analyze the interpretability of deep learning mortality prediction models and observe that (1) the best-performing interpretability method successfully identifies critical features for mortality prediction on various prediction models as well as recognizing new important features that domain knowledge does not consider; (2) prediction models rely on demographic features, raising concerns in fairness. Therefore, we then evaluate the fairness of models and do observe the unfairness: (1) there exists disparate treatment in prescribing mechanical ventilation among patient groups across ethnicity, gender and age; (2) models often rely on racial attributes unequally across subgroups to generate their predictions. We further draw concrete connections between interpretability methods and fairness metrics by showing how feature importance from interpretability methods can be beneficial in quantifying potential disparities in mortality predictors. Our analysis demonstrates that the prediction performance is not the only factor to consider when evaluating models for healthcare applications, since high prediction performance might be the result of unfair utilization of demographic features. Our findings suggest that future research in AI models for healthcare applications can benefit from utilizing the analysis workflow of interpretability and fairness as well as verifying if models achieve superior performance at the cost of introducing bias. [ABSTRACT FROM AUTHOR]

The introduction of digital currency electronic payment (DCEP) by the Central Bank of China is conducive to the central bank's timely grasp of macroeconomic dynamics and the internationalization of RMB. As DCEP is one of the first digital currencies issued by the central bank to be used on a large scale internationally, it is necessary to conduct research on its user adoption. Therefore, this research extends the unified theory of acceptance and use of technology (UTAUT) to explore factors affecting the adoption of DCEP. The researchers cooperated with city banks that have started to use DCEP, and distributed questionnaires to users in the lobbies of these banks. A total of 295 valid questionnaires were empirically examined with Smart-PLS. The results indicate that perceived fairness, habits, social influence and national identity have significant effects on usage, with p values less than 0.05. National identity is shown to be a significant moderator of the relationships between perceived fairness, habit, perceived risk and usage, with p values less than 0.05. National identity is shown to have no moderating effect between social influence and usage, with a p value greater than 0.05. This research provides the central bank and the government with suggestions to increase user enthusiasm and reduce user perceived risks, thereby promoting the widespread use of DCEP. [ABSTRACT FROM AUTHOR]

Ingroup favoritism and fairness are two potentially competing motives guiding intergroup behaviors in human. Here, we investigate if and how limited resources can modulate the way these two motives affect individuals' decisions in intergroup situation. In the present study, participants (N = 58) were asked to accept or reject three types of resource allocation proposals generated by a computer: the ingroup advantageous condition, outgroup advantageous condition, and neutral condition. In general, participants were more willing to accept the proposals in the ingroup advantageous condition than the outgroup advantageous or the neutral conditions, and also in the moderate inequality than the extreme inequality condition. This may indicate that people sought a careful balance between ingroup favoritism and fairness, although we also found marked individual differences in their preferences for ingroup favoritism or fairness. Importantly, as predicted, participants were more likely to show ingroup favoritism only when limited resources affect the well-being of ingroup members. The present study provides novel insights into the situational and personality factors affecting human intergroup behaviors, shedding light on motives underlying intergroup conflicts prevalent in human societies. [ABSTRACT FROM AUTHOR]