$$\mathsf {FABBOO}$$ - Online Fairness-Aware Learning Under Class Imbalance

Iosifidis, Vasileios; Ntoutsi, Eirini

doi:10.1007/978-3-030-61527-7_11

Cited by 17 publications

(17 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our contributions are summarized as follows: i) we propose FABBOO, a fairness-and class imbalance-aware boosting approach that is able to tackle class imbalance as well as mitigate different parity-based discriminatory outcomes, ii) we introduce the notion of cumulative fairness in streams, which accounts for cumulative discriminatory outcomes, iii) our experiments, in a variety of real-world and synthetic datasets, show that our approach outperforms existing fairness-aware approaches. This work is an extension of our previous work [23]. The major changes include: i) modifying the distribution update part by also reducing the majority weights, ii) extending FABBOO to facilitate another parity-based notion of fairness, namely predictive equality, iii) adding two real-world datasets to the experimental evaluation, iv) adding a recently published state-of-the-art imbalance-aware stream classifier [2] for comparison, iv) providing a detailed analysis w.r.t FABBOO's hyper-parameters selection.…”

mentioning

confidence: 68%

Online Fairness-Aware Learning with Imbalanced Data Streams

Iosifidis,

Zhang,

Ntoutsi

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Data-driven learning algorithms are employed in many online applications, in which data become available over time, like network monitoring, stock price prediction, job applications, etc. The underlying data distribution might evolve over time calling for model adaptation as new instances arrive and old instances become obsolete. In such dynamic environments, the so-called data streams, fairness-aware learning cannot be considered as a one-off requirement, but rather it should comprise a continual requirement over the stream. Recent fairness-aware stream classifiers ignore the problem of class imbalance, which manifests in many real-life applications, and mitigate discrimination mainly because they "reject" minority instances at large due to their inability to effectively learn all classes.In this work, we propose FABBOO, an online fairness-aware approach that maintains a valid and fair classifier over the stream. FABBOO is an online boosting approach that changes the training distribution in an online fashion by monitoring stream's class imbalance and tweaks its decision boundary to mitigate discriminatory outcomes over the stream. Experiments on 8 real-world and 1 synthetic datasets from different domains with varying class imbalance demonstrate the superiority of our method over state-of-the-art fairnessaware stream approaches with a range (relative) increase [11.2%-14.2%] in balanced accuracy, [22.6%-31.8%] in gmean, [42.5%-49.6%] in recall, [14.3%-25.7%] in kappa and [89.4%-96.6%] in statistical parity (fairness).

show abstract

mentioning

confidence: 68%

Online Fairness-Aware Learning with Imbalanced Data Streams

Iosifidis,

Zhang,

Ntoutsi

2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Previous researches consider sex as a protected attribute (Iosifidis & Ntoutsi, 2019; Iosifidis & Ntoutsi, 2020; Ristanoski et al, 2013). Attribute race = { white , black , asian‐pac‐islander , amer‐indian‐eskimo , other } could be also employed as a protected attribute because it has the same role as in the original Adult dataset.…”

Section: Datasets For Fairnessmentioning

confidence: 99%

A survey on datasets for fairness‐aware machine learning

Quy

Roy

Iosifidis

et al. 2022

WIREs Data Min & Knowl

Self Cite

View full text Add to dashboard Cite

As decision‐making increasingly relies on machine learning (ML) and (big) data, the issue of fairness in data‐driven artificial intelligence systems is receiving increasing attention from both research and industry. A large variety of fairness‐aware ML solutions have been proposed which involve fairness‐related interventions in the data, learning algorithms, and/or model outputs. However, a vital part of proposing new approaches is evaluating them empirically on benchmark datasets that represent realistic and diverse settings. Therefore, in this paper, we overview real‐world datasets used for fairness‐aware ML. We focus on tabular data as the most common data representation for fairness‐aware ML. We start our analysis by identifying relationships between the different attributes, particularly with respect to protected attributes and class attribute, using a Bayesian network. For a deeper understanding of bias in the datasets, we investigate interesting relationships using exploratory analysis. This article is categorized under: Commercial, Legal, and Ethical Issues > Fairness in Data Mining Fundamental Concepts of Data and Knowledge > Data Concepts Technologies > Data Preprocessing

show abstract

“…Applying generalization methods by training a model that is robust to various distribution shifts, researchers examine the effects of robustness on fairness. Iosifidis et al [25,26] apply pre-processing and distribution shifts methods in streaming classification framework in which the dataset or the model has to be corrected at each time step so the model stays stable. Singh et al [44] utilize causal learning to build a model that is insensitive to distribution shifts that might occur in the features (i.e., covariate shift).…”

Section: Related Workmentioning

confidence: 99%

ABCinML: Anticipatory Bias Correction in Machine Learning Applications

Almuzaini

Bhatt

Pennock

et al. 2022

2022 ACM Conference on Fairness, Accountability, and Transparency

View full text Add to dashboard Cite

The idealization of a static machine-learned model, trained once and deployed forever, is not practical. As input distributions change over time, the model will not only lose accuracy, any constraints to reduce bias against a protected class may fail to work as intended. Thus, researchers have begun to explore ways to maintain algorithmic fairness over time. One line of work focuses on dynamic learning: retraining after each batch, and the other on robust learning which tries to make algorithms robust against all possible future changes. Dynamic learning seeks to reduce biases soon after they have occurred and robust learning often yields (overly) conservative models. We propose an anticipatory dynamic learning approach for correcting the algorithm to mitigate bias before it occurs. Specifically, we make use of anticipations regarding the relative distributions of population subgroups (e.g., relative ratios of male and female applicants) in the next cycle to identify the right parameters for an importance weighing fairness approach. Results from experiments over multiple real-world datasets suggest that this approach has promise for anticipatory bias correction. CCS CONCEPTS• Computing methodologies → Machine learning; Batch learning.

show abstract

$$\mathsf {FABBOO}$$ - Online Fairness-Aware Learning Under Class Imbalance

Cited by 17 publications

References 20 publications

Online Fairness-Aware Learning with Imbalanced Data Streams

Online Fairness-Aware Learning with Imbalanced Data Streams

A survey on datasets for fairness‐aware machine learning

ABCinML: Anticipatory Bias Correction in Machine Learning Applications

Contact Info

Product

Resources

About