Algorithmic Bias: From Discrimination Discovery to Fairness-aware Data Mining

By Sara Hajian, Francesco Bonchi, Carlos Castillo. 2016

394 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Health Informatics, Safety Research, AI Ethics, Fairness & Society

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Health Informatics

In Safety Research

In AI Ethics, Fairness & Society

Citations in the corpus

Conscientious Classification: A Data Scientist's Guide to Discrimination-Aware Classification (2017, positive): “29 This is in spite of the fact that, as can be seen in the empirical studies we cite, 12,17,27 differences between the highest performing state of the art and previously established methods are often within a few percentage points of each other.”
Social Data: Biases, Methodological Pitfalls, and Ethical Boundaries (2019, background): “This problem is often referred to as algorithmic discrimination or algorithmic bias (see, e.g., Hajian et al., 2016).”
Stable and Fair Classification (2019, methodology): “Recent studies (Hajian et al., 2016; 2015; Kashid et al., 2015; 2017; Ruggieri et al., 2014) have expanded the application of methods to achieve both goals; see a recent paper (Ekstrand et al.”
Assessing gender bias in machine translation: a case study with Google Translate (2018, methodology): “nt compromise in the performance of the word embedding technique. They are not alone: there is a growing eort to systematically discover and resolve issues of algorithmic bias in black-box algorithms[18]. The success of these results suggest that a similar technique could be used to remove gender bi”
Increasing Trust in AI Services through Supplier's Declarations of Conformity (2018, background): “Fairness AI fairness is a rapidly growing topic of inquiry [35].”
FAHT: An Adaptive Fairness-aware Decision Tree Classifier (2019, methodology): “A number of research approaches have been proposed to address the problem of bias and discrimination in machine learning systems due to the inherent bias in data and the complex interaction between data and learning algorithms [Pe-dreshi et al. , 2008; Zliobaite, 2015; Hajian et al. , 2016].”
Clinical applications of machine learning in cardiovascular disease and its relevance to cardiac imaging. (2018, background): “93 While there have been no such reported incidents in the healthcare setting, further work is necessary for developing methodologies to rule out clinically non-relevant biases before a machine’s clinical decision-making can be trusted.”
Explainable AI: Interpreting, Explaining and Visualizing Deep Learning (2019, background): “But also anti-discrimination and fairness aspects have been widely discussed in the context of AI [19,28].”
Non-Discriminatory Machine Learning through Convex Fairness Criteria (2018, background): “There has also been a significant amount of work in computational social science towards achieving fairness in data collection, social networks etc (Olteanu et al. 2016),(Tufekci 2014),(Morstatter and Liu 2017),(Hajian, Bonchi, and Castillo 2016).”
Activism by the AI Community: Analysing Recent Achievements and Future Prospects (2020, background): “AI deployment is already influencing markets and the economy, justice and distributive decisions by governments, and our elections [32, 63].”
Fairness without Demographics through Adversarially Reweighted Learning (2020, methodology): “There is also a large body of work on incorporating these fairness notions into ML models, including learning better representations [56, 38, 37] and adding fairness constraints in the learning objective [19, 55, 49, 31], through post-processing the decisions [20], through adversarial learning [57, ”
Interpretable Machine Learning in Healthcare (2018, background): “These predictive algorithms are having disruptive effects on society [32] and resulting in unforeseen consequences [12] like deskilling of physicians.”
Trends and Trajectories for Explainable, Accountable and Intelligible Systems: An HCI Research Agenda (2018, methodology): “Hajian called for data mining systems which are discrimination-con-scious by-design [76]; such as fair systems for loan risk prediction [141], and bail flight risk of violent crime suspects [29], as well as methods to detect bias in black box models, such as Quantitative Influence (QI) [38] and…”
FA*IR: A Fair Top-k Ranking Algorithm (2017, background): “ithm. Section 5 presents experimental results. Section 6 presents our conclusions and future work. 2 RELATED WORK Anti-discrimination has only recently been considered from an algorithmic perspective [15]. Some proposals are oriented to discovering and measuring discrimination (e.g., [1, 3, 31]); wh”
Multistakeholder recommendation: Survey and research directions (2020, background): “services may be unfairly underrated. 7.2 Review of Recommender Fairness Fairness in algorithmic systems admits of a variety of denitions and has been the subject of considerable recent research. See [58] for a survey of fairness in data mining applications. Fairness in recommender systems is a subtl”
Addressing multiple metrics of group fairness in data-driven decision making (2020, background): “However, a growing literature has uncovered algorithmic discrimination in sensitive contexts and described fairness-aware ML algorithms [4,17].”