Fairlearn is an open-source Python toolkit for assessing and improving the fairness of machine-learning models with respect to sensitive attributes such as race, gender, or age. The library combines a disaggregated evaluation framework, several bias-mitigation algorithms, and a set of benchmark datasets, all designed to plug into the scikit-learn fit/predict/transform API [6]. The project was started in May 2018 by Microsoft Research as a small implementation accompanying the Agarwal et al. ICML paper A Reductions Approach to Fair Classification [3], expanded into a broader toolkit during 2019 and 2020, and moved to independent community governance in 2021. It is released under the MIT licence and is developed in the open at fairlearn/fairlearn on GitHub, where it has more than 2,300 stars, with a documentation site at fairlearn.org [7]. The current release is version 0.14.0, published on 7 June 2026, and the PyPI package is downloaded roughly 185,000 times per month [26][27].

Alongside AIF360, Fairlearn is one of the two most widely cited open-source toolkits for algorithmic fairness [20]. Its central technical contribution is a clean implementation of the reductions framework for fair classification, in which a fairness-constrained learning problem is reduced to a sequence of standard cost-sensitive classification problems that any black-box classifier can solve [3]. This reductions approach gave Fairlearn its name and remains the design idea that distinguishes the library from competing toolkits that focus more on cataloguing methods than on a single coherent framework.

History

Fairlearn began in May 2018 as a small Python package open-sourced by Miroslav Dudík at Microsoft Research to accompany "A Reductions Approach to Fair Classification" by Agarwal, Beygelzimer, Dudík, Langford, and Wallach, presented at the 35th International Conference on Machine Learning (ICML 2018) in Stockholm [3]. The first release was narrow: a single ExponentiatedGradient reduction wrapping any binary classifier under demographic parity or equalised odds. It was essentially a reference implementation that let readers reproduce the paper's experiments.

During the second half of 2019 and into 2020 the project was expanded by Microsoft Research and Azure Machine Learning teams. The major additions: a metrics module built around MetricFrame, a ThresholdOptimizer postprocessing algorithm implementing Hardt-Price-Srebro (2016) [4], a second reductions algorithm GridSearch based on Agarwal-Dudík-Wu (ICML 2019, "Fair Regression: Quantitative Definitions and Reduction-Based Algorithms") [5], and a Jupyter dashboard widget (later deprecated in favour of the Responsible AI Toolbox).

The state of the project at the end of this period was documented in Bird, Dudík, Edgar, Horn, Lutz, Milan, Sameki, Wallach, and Walker (2020), "Fairlearn: A toolkit for assessing and improving fairness in AI", Microsoft Research Technical Report MSR-TR-2020-32, May 2020 [1]. That whitepaper covers the API up through version 0.4.x and is still cited as the canonical Fairlearn reference.

In 2021 the project moved from Microsoft to neutral governance. The GitHub organisation fairlearn was created, the repository was migrated from microsoft/fairlearn to fairlearn/fairlearn, and a steering committee structure was set up under a separate fairlearn/governance repository [8]. Microsoft remained a funder of maintainer work alongside Eindhoven University of Technology and Hugging Face [6].

In 2023, Hilde Weerts, Miroslav Dudík, Richard Edgar, Adrin Jalali, Roman Lutz, and Michael Madaio published "Fairlearn: Assessing and Improving Fairness of AI Systems" in the Journal of Machine Learning Research (volume 24, paper 257) [2]. It serves as an updated reference after the governance transition, documenting the adversarial mitigation, CorrelationRemover preprocessing, and the modern MetricFrame-based assessment workflow. As of July 2026 the library is at version 0.14.0 (released 7 June 2026), the successor to 0.13.0 (19 October 2025) [27].

Year	Milestone
2018 (May)	First release as a small reductions package, accompanying Agarwal et al. ICML 2018
2019	Expanded to include metrics, GridSearch, and ThresholdOptimizer; Azure ML team contributions
2020 (May)	Bird et al. Microsoft Research Technical Report MSR-TR-2020-32 published
2020	Visualisation dashboard widget added (later deprecated)
2021	Project transitioned to community governance under `fairlearn` GitHub organisation
2022	`MetricFrame`-centric API stabilised; deprecation of legacy classes
2023	Weerts et al. paper published in Journal of Machine Learning Research
2024	Continued releases adding datasets, adversarial mitigation, new metrics
2025 (Oct)	Version 0.13.0 adds `PrototypeRepresentationLearner` preprocessing (Zemel et al. 2013)
2026 (Jun)	Version 0.14.0 released: Python 3.13, scikit-learn 1.5+, Narwhals dataframe support

Is Fairlearn still a Microsoft project?

No. Fairlearn adopted neutral governance in 2021 and describes itself as "completely community-driven" under the Fairlearn Organization's project governance [6][8]. The library is still funded in part by Microsoft, which has paid several maintainers (Miroslav Dudík, Richard Edgar, Roman Lutz, and Michael Madaio) since the project began in 2018, but the funding base is now shared: Eindhoven University of Technology has funded Hilde Weerts since March 2020, and Hugging Face has funded Adrin Jalali since February 2022 [6]. The maintainer team listed on the project's About page in 2026 includes Adrin Jalali, Hilde Weerts, Michael Madaio, Miroslav Dudík, Richard Edgar, Roman Lutz, Allie Saizan, Tamara Atanasoska, and Tahar Allouche [6].

What are Fairlearn's goals and design philosophy?

The Fairlearn documentation is unusually explicit about what the library is and is not. The maintainers state that fairness is a sociotechnical problem and cannot be reduced to a metric you compute and a constraint you enforce, because what counts as fair depends on social context and on whose harm matters. As the project's About page puts it, "fairness in AI systems is a sociotechnical challenge", and "it is not possible to fully 'debias' a system or to guarantee fairness" [6]. The library is positioned as a tool that helps a human decide where to look, not an automated debiaser. This framing echoes interview research with industry practitioners, who report needing support across the whole fairness workflow rather than a one-click fix (Holstein et al. 2019) [16].

The project favours harm-based language over the words bias and debiasing. The 2020 whitepaper distinguishes between allocation harms, in which a model affects who gets a resource or opportunity (a loan, a job interview, parole), and quality-of-service harms, in which a model performs differently for different groups (a speech recogniser with higher word-error rate on Black speakers, for example) [1]. A third principle, present in the code rather than the marketing material, is scikit-learn compatibility: almost every Fairlearn class follows the scikit-learn estimator interface, so a fairness-constrained classifier drops into an existing pipeline alongside StandardScaler, OneHotEncoder, and GridSearchCV with no special glue. The conceptual material in the user guide draws on the standard reference text, Barocas, Hardt, and Narayanan's Fairness and Machine Learning: Limitations and Opportunities [11].

How is Fairlearn structured?

Fairlearn is organised as a single Python package, fairlearn, with a small set of submodules.

MetricFrame and the metrics module

fairlearn.metrics.MetricFrame is the centrepiece of the assessment side of the library. It takes one or more performance metrics (any scikit-learn metric or custom callable), the true labels, the predicted labels, and a sensitive feature vector, and returns a pandas DataFrame of metric values disaggregated by group. The same object computes summary statistics across groups including difference, ratio, group_min, and group_max. Because MetricFrame accepts arbitrary metrics, it is also the recommended way to compute custom or domain-specific fairness measures. The metrics module also exposes scalar wrappers like demographic_parity_difference, demographic_parity_ratio, equalized_odds_difference, and equalized_odds_ratio.

Reductions module

fairlearn.reductions contains the in-processing mitigation algorithms based on the reductions framework. The two core classes are ExponentiatedGradient, an iterative algorithm that approximately solves a saddle-point problem over Lagrange multipliers and re-weights training samples each iteration; and GridSearch, which sweeps a grid of Lagrange multipliers and trains one classifier per grid point, returning a list of trade-off solutions [3]. The implementation supports demographic parity, equalised odds, TPR parity, FPR parity, error rate parity, and bounded group loss. Both classes wrap any scikit-learn-compatible classifier or regressor as the base estimator, with the user providing a Moment object from fairlearn.reductions that defines the fairness constraint mathematically.

Postprocessing module

fairlearn.postprocessing.ThresholdOptimizer implements the Hardt-Price-Srebro 2016 method [4]. Given an existing scored classifier and the sensitive attribute on held-out data, it computes a per-group decision threshold (or a randomised mixture of thresholds) such that the resulting classifier satisfies the chosen fairness constraint while remaining optimal for accuracy. It supports demographic parity, equalised odds, true positive rate parity, and false positive rate parity. The class is the right tool when retraining is expensive or impossible: it operates on classifier scores rather than the model itself.

Preprocessing and adversarial modules

fairlearn.preprocessing provides two preprocessing transforms. CorrelationRemover removes linear correlation between sensitive features and non-sensitive features, leaving a transformed feature matrix any standard learner can use; an alpha parameter controls the strength of the filtering, with alpha=1.0 removing all measured linear correlation. Version 0.13.0 (October 2025) added PrototypeRepresentationLearner, an implementation of the learning-fair-representations method of Zemel, Wu, Swersky, Pitassi, and Dwork (2013) [25]. It learns a latent representation of the data using a set of prototype vectors and a stochastic mapping, minimising a three-term objective that trades off reconstruction error, classification error, and an approximation of the demographic parity difference [25]. Even with this addition, Fairlearn implements fewer preprocessing methods than AIF360, so users who need richer preprocessing are still encouraged to combine Fairlearn metrics with AIF360 preprocessing transforms [12]. fairlearn.adversarial.AdversarialFairnessClassifier and AdversarialFairnessRegressor, added in later releases, implement adversarial debiasing in which a predictor and an adversary are trained jointly so the adversary cannot recover the sensitive attribute from the predictor's outputs (Zhang, Lemoine, and Mitchell 2018) [22]. This is a PyTorch-backed in-processing method that fits with deep-learning pipelines.

Datasets module

fairlearn.datasets provides loader functions for benchmark datasets used in the fairness literature.

Function	Dataset	Task	Sensitive features	Notes
`fetch_adult`	Adult / Census Income	Binary classification	Sex, race	UCI; income > 50,000 USD; most cited fairness benchmark
`fetch_diabetes_hospital`	Diabetes 130-Hospitals	Binary classification	Race, gender	1999 to 2008 hospital admissions; readmission
`fetch_acs_income`	ACSIncome	Regression	Race, sex	Folktables 2018 ACS, 1.6M records
`fetch_bank_marketing`	UCI Bank Marketing	Binary classification	Age	Portuguese telemarketing campaign
`fetch_credit_card`	UCI credit card default	Binary classification	Sex, age	Taiwan 2005

The ACSIncome loader, contributed in the 0.7 series, is based on Ding et al. (2021)'s Folktables data, designed in response to the limitations of the original Adult dataset [19].

Fairlearn is built on numpy, pandas, and scikit-learn. The reductions and postprocessing classes accept any object with fit and predict methods, so PyTorch and TensorFlow models can be used as base estimators if wrapped in a thin scikit-learn-compatible class. The library has no deep-learning-specific module beyond the adversarial mitigation, and most usage in practice is on tabular data.

What fairness metrics does Fairlearn provide?

Fairlearn exposes a focused, curated set of fairness metrics. Unlike AIF360 (more than seventy metrics, many redundant), Fairlearn keeps the core set small and relies on MetricFrame for everything else [12].

Metric	What it measures	Origin
Demographic parity difference	Maximum difference in selection rate between groups	Calders & Verwer 2010; Dwork et al. 2012
Demographic parity ratio	Ratio of minimum to maximum selection rate; comparable to the four-fifths rule for disparate impact	US EEOC Uniform Guidelines, 1978
Equalised odds difference	Maximum of TPR difference and FPR difference between groups	Hardt, Price, Srebro 2016
Equalised odds ratio	Minimum-to-maximum ratio of TPR and FPR across groups	Hardt, Price, Srebro 2016
Equal opportunity (TPR parity)	Equality of true positive rates across groups	Hardt, Price, Srebro 2016
False positive rate parity	Equality of false positive rates across groups	Hardt, Price, Srebro 2016
Selection rate	Fraction of positive predictions per group	Definitional
Disaggregated sklearn metrics	Accuracy, precision, recall, F1, MAE, MSE, etc., by group	Computed via `MetricFrame`
Worst-case accuracy	Lowest accuracy among any group	Hashimoto et al. 2018 [21]

A distinguishing feature is that Fairlearn does not ship a single canonical fairness metric and instead encourages the user to choose. The user guide walks through Chouldechova (2017) and Kleinberg, Mullainathan, & Raghavan (2017), the canonical incompatibility of fairness metrics papers, to make clear that picking a metric is a value judgement [9][10]. Fairlearn also supports fairness metrics for regression, including bounded group loss, which asks that prediction error within any sensitive group stay below a chosen level.

How does Fairlearn mitigate bias?

Fairlearn organises mitigation algorithms by where they intervene in the pipeline, the taxonomy made standard by Friedler et al. (2019) [15]. The library is less encyclopaedic than AIF360 and contains a smaller, curated set of methods, with reductions as the design centre.

Category	Algorithm	Original paper	Supported constraints
Preprocessing	`CorrelationRemover`	Fairlearn user guide	Linear correlation removal between sensitive and non-sensitive features
Preprocessing	`PrototypeRepresentationLearner`	Zemel, Wu, Swersky, Pitassi, Dwork 2013	Learns a fair latent representation (approximate demographic parity)
In-processing (reductions)	`ExponentiatedGradient`	Agarwal, Beygelzimer, Dudík, Langford, Wallach 2018	DemographicParity, EqualizedOdds, TPR/FPR parity, ErrorRateParity, BoundedGroupLoss
In-processing (reductions)	`GridSearch`	Agarwal et al. 2018 §3.4; Agarwal, Dudík, Wu 2019 (regression)	Same as `ExponentiatedGradient`
In-processing (adversarial)	`AdversarialFairnessClassifier`, `AdversarialFairnessRegressor`	Zhang, Lemoine, Mitchell 2018	Demographic parity (via adversary loss)
Postprocessing	`ThresholdOptimizer`	Hardt, Price, Srebro 2016	DemographicParity, EqualizedOdds, TPR/FPR parity

How does the reductions framework work?

The reductions framework is Fairlearn's central technical contribution. The starting observation, due to Agarwal et al. (2018), is that any fairness constraint that can be written as a linear inequality on the conditional moments of a classifier (demographic parity, equalised odds, equal opportunity, or any conjunction) defines a constrained empirical risk minimisation problem [3]. Its Lagrangian dual can be expressed as a saddle-point game between the classifier and a vector of Lagrange multipliers, and the inner minimisation, for any fixed multiplier vector, is a standard cost-sensitive classification problem with sample weights derived from the multipliers. So any black-box cost-sensitive classifier can be plugged in as an oracle to solve the fair-classification problem without modification. The authors describe the design in one sentence: "The key idea is to reduce fair classification to a sequence of cost-sensitive classification problems, whose solutions yield a randomized classifier with the lowest (empirical) error subject to the desired constraints" [3].

ExponentiatedGradient implements this with the no-regret online learning algorithm of Freund and Schapire, which converges to the optimal classifier in O(1/sqrt(T)) iterations. GridSearch is a simpler alternative that enumerates a grid of multiplier values, trains one classifier per grid point, and returns the Pareto frontier of accuracy-fairness trade-offs.

The practical advantages are significant. The user can swap in any base learner without changing the fairness algorithm. The same code path supports any fairness constraint cast as linear inequalities on conditional moments, which covers most of the group-fairness literature. The output is a randomised classifier (a distribution over deterministic classifiers), necessary in general because deterministic classifiers cannot always satisfy demographic parity or equalised odds exactly. The algorithm also comes with theoretical guarantees on convergence and suboptimality. The extension to fair regression in Agarwal, Dudík, and Wu (2019) replaces binary outcomes with real-valued ones under Lipschitz-continuous losses, a capability AIF360 does not have natively [5].

Postprocessing with `ThresholdOptimizer`

The Hardt-Price-Srebro (2016) postprocessing method takes a fully trained classifier that produces real-valued scores and learns a per-group decision rule to satisfy a fairness constraint [4]. For demographic parity it picks per-group thresholds so the positive-prediction rate is equal across groups. For equalised odds it can pick a randomised mixture of two thresholds per group, because deterministic thresholds cannot match both TPR and FPR exactly except by coincidence. The optimisation is convex (a linear program per group) and the implementation is fast, which is why ThresholdOptimizer is often the first thing practitioners try.

How does Fairlearn compare to AIF360 and other fairness toolkits?

Fairlearn is one of several open-source fairness toolkits that emerged in the 2018 to 2020 window. The libraries differ in licence, language, focus, and design philosophy, and many production teams use more than one.

Toolkit	Organisation	Year	Language	Licence	Focus	Mitigation
Fairlearn	Microsoft, then community	2018 (May)	Python	MIT	Reductions + sklearn-style assessment	6 classes (2 pre, 3 in, 1 post)
AI Fairness 360 (AIF360) [12]	IBM, then LF AI & Data	2018 (Sept)	Python, R	Apache 2.0	Encyclopaedic catalogue	13+ across pre, in, post
AI Explainability 360 (AIX360)	IBM, LF AI & Data	2019	Python	Apache 2.0	Sister project for explainability	None (explanation only)
Aequitas [13]	University of Chicago CDSPP	2018	Python, CLI	MIT	Bias audit reports	None (audit only)
TF Fairness Indicators [14]	Google, TensorFlow team	2019 (Dec)	Python (TFX)	Apache 2.0	Slice-based fairness for TF models	None (evaluation only)
What-If Tool	Google PAIR	2018	Python (TensorBoard)	Apache 2.0	Interactive what-if exploration	None
Themis-ML	Niels Bantilan (academic)	2017	Python	MIT	Predates AIF360 and Fairlearn	A few preprocessing methods
OxonFair [24]	University of Oxford	2024	Python	MIT	Flexible postprocessing	Postprocessing-focused

Fairlearn and AIF360 are the most direct rivals because both cover metrics and mitigation across pre, in, and post-processing.

Aspect	Fairlearn	AIF360
Number of mitigation algorithms	Smaller, more curated	Encyclopaedic catalogue
In-processing approach	Reductions framework as core abstraction	Diverse stand-alone algorithms
Regression support	Yes (GridSearch with BoundedGroupLoss; ACSIncome)	Limited
More than two sensitive groups	Supported throughout	Most methods assume binary protected attribute
Language bindings	Python only	Python and R
API style	scikit-learn-style fit/predict/transform	sklearn-style for `aif360.sklearn`, otherwise its own

Lee and Singh (2021) compared the user experience of these toolkits with practitioners and found Fairlearn easier to use for newcomers, while AIF360 had broader algorithmic coverage [20]. A common combined pattern is to use AIF360 preprocessing (Reweighing, LearningFairRepresentations) on raw data, then Fairlearn's MetricFrame and ThresholdOptimizer on the predictions.

How do you use Fairlearn in practice?

The scikit-learn-compatible API is the dominant aspect of working with Fairlearn day to day. A typical training loop looks like this:

from fairlearn.reductions import ExponentiatedGradient, DemographicParity
from sklearn.linear_model import LogisticRegression

base = LogisticRegression(solver='liblinear')
mitigator = ExponentiatedGradient(base, constraints=DemographicParity())
mitigator.fit(X_train, y_train, sensitive_features=A_train)
y_pred = mitigator.predict(X_test)

The only fairness-specific arguments are constraints and sensitive_features. The same shape works for GridSearch, ThresholdOptimizer, and CorrelationRemover. MetricFrame follows pandas conventions and integrates with notebooks and existing reporting code. Visualisation helpers built on matplotlib and seaborn ship with the library.

On the Microsoft side, Fairlearn is integrated into the Azure Machine Learning Responsible AI dashboard (as one of four pillars alongside interpretability, error analysis, and counterfactuals) and into the open-source Responsible AI Toolbox at microsoft/responsible-ai-toolbox [23]. The deprecated Fairlearn dashboard widget was effectively absorbed into the Toolbox, leaving the core library focused on algorithms. Documentation at fairlearn.org includes a long-form user guide, API reference, and Jupyter notebook examples [6]. The user guide spends as much space on fairness concepts (Chouldechova's impossibility result, allocation versus quality-of-service harms, intersectional fairness) as on API details.

Who uses Fairlearn?

Fairlearn is cited in academic ML fairness courses, including Berkeley's CS 294 and Carnegie Mellon's machine learning fairness course, alongside AIF360. The library is used inside Microsoft as part of Responsible AI requirements for high-impact ML systems and is a technical component of Azure Machine Learning's Responsible AI offering for enterprise customers.

As of July 2026 the GitHub repository has more than 2,300 stars and over 500 forks (504), with more than 1,000 commits (1,041 on the main branch) and an active maintainer team [7]. The PyPI package is downloaded roughly 185,000 times per month (about 184,800 downloads in the 30 days to early July 2026, according to PyPI Stats) [26]. The library has been integrated into MLflow examples for fairness logging and into Hugging Face's evaluate library for fairness metrics. Fairlearn has been referenced in technical annexes of model risk management documents and in submissions discussing fairness requirements under the EU AI Act, the New York City automated employment decision tool law, and the Colorado AI Act. The library provides measurements and interventions a compliance team can incorporate into a wider audit process; it does not provide regulatory compliance by itself.

What are Fairlearn's limitations?

Several limitations follow from the field rather than from the library specifically. Chouldechova (2017) and Kleinberg, Mullainathan, & Raghavan (2017) proved that calibration, equal false-positive rates, and equal false-negative rates cannot all be satisfied simultaneously when base rates differ across groups, except in degenerate cases [9][10]. Fairlearn implements the metrics on either side of the trade-off but does not pretend the conflict can be resolved.

Fairlearn's mitigation algorithms target group fairness criteria (demographic parity, equalised odds, and similar). The library does not natively implement individual fairness (Dwork et al. 2012) [17] or counterfactual fairness (Kusner et al. 2017) [18]. Most metrics and all mitigation algorithms require the sensitive attribute to be observed in the data, which may be legally restricted (under GDPR rules for race or health data) or unavailable. Fairness through unawareness, which omits the sensitive attribute from the model, is known to be insufficient because protected attributes can be reconstructed from correlated features.

ExponentiatedGradient trains the base classifier many times (often 50 to 200 iterations), which is slow if the base classifier is itself expensive; GridSearch is faster but explores a coarser frontier, and ThresholdOptimizer is cheapest. While MetricFrame supports intersectional disaggregation, the in-processing mitigation algorithms typically treat the sensitive feature as a single categorical variable, so rich intersectional fairness in the sense of Kearns et al. (2018)'s GerryFair classifier is not part of Fairlearn. Although the adversarial mitigation classes work with PyTorch backbones, the bulk of Fairlearn's tooling is tabular; image and text fairness, especially in large pretrained models, is largely outside the scope of the library.

Recent developments (2025 to 2026)

Releases have continued at one or two minor versions per year. The 0.10 and 0.11 releases added more Moment classes for the reductions framework, including better support for non-binary sensitive features, and stabilised the adversarial fairness classes. Version 0.12 (December 2024) and version 0.13.0 (19 October 2025) tightened scikit-learn 1.5+ compatibility and improved typing; 0.13.0 additionally introduced the PrototypeRepresentationLearner preprocessing method, added support for relaxed constraints in ThresholdOptimizer, and began integrating the Narwhals library for dataframe-agnostic code [25][27]. Version 0.14.0, released on 7 June 2026, dropped Python 3.9 and added Python 3.13, raised the minimum scikit-learn to 1.5.0, pinned the minimum PyTorch to 2.8.0, added scipy 1.16 compatibility, and extended Narwhals support to the postprocessing module and the GroupFeature class; it also fixed handling of degenerate sensitive-feature values in PrototypeRepresentationLearner.fit and made GridSearch.fit return self per the scikit-learn convention [27].

The project's positioning has shifted from "the Microsoft fairness library" toward "the community fairness library that Microsoft funds". Maintainer funding now comes from Microsoft, Eindhoven University of Technology, and Hugging Face [6]. In the broader landscape, OxonFair (Oxford, 2024) added flexible postprocessing [24], AIF360 added MDSS bias scanning and intersectional metrics, and Hugging Face's evaluate library now exposes Fairlearn and AIF360 metrics. The trend is to combine fairness measurement with model documentation (Datasheets for Datasets, Model Cards), interpretability, and privacy in a unified responsible-AI workflow.

Why is Fairlearn significant?

Alongside AI Fairness 360 (AIF360), Fairlearn is one of the two leading open-source toolkits that took the academic fairness literature out of conferences and into mainstream Python machine-learning practice [12]. Either is a reasonable default for a team adding fairness checks to a production system. Fairlearn also brought the reductions approach into widespread practice. The Agarwal et al. 2018 paper would still be cited without the library, but it would not be running in compliance pipelines at banks or in Azure Machine Learning [3]. The library demonstrated that fairness mitigation can be a small, modular component that integrates cleanly with standard ML pipelines, a design choice that has shaped how subsequent fairness toolkits including OxonFair and the Hugging Face evaluate fairness modules have positioned themselves [24].

See also

References

Bird, S., Dudík, M., Edgar, R., Horn, B., Lutz, R., Milan, V., Sameki, M., Wallach, H., & Walker, K. (2020). "Fairlearn: A toolkit for assessing and improving fairness in AI." Microsoft Research Technical Report MSR-TR-2020-32. https://www.microsoft.com/en-us/research/publication/fairlearn-a-toolkit-for-assessing-and-improving-fairness-in-ai/
Weerts, H., Dudík, M., Edgar, R., Jalali, A., Lutz, R., & Madaio, M. (2023). "Fairlearn: Assessing and Improving Fairness of AI Systems." Journal of Machine Learning Research, 24(257):1-8. arXiv:2303.16626. https://jmlr.org/papers/v24/23-0389.html
Agarwal, A., Beygelzimer, A., Dudík, M., Langford, J., & Wallach, H. (2018). "A Reductions Approach to Fair Classification." ICML 2018. arXiv:1803.02453.
Hardt, M., Price, E., & Srebro, N. (2016). "Equality of Opportunity in Supervised Learning." NeurIPS 2016. arXiv:1610.02413.
Agarwal, A., Dudík, M., & Wu, Z. S. (2019). "Fair Regression: Quantitative Definitions and Reduction-Based Algorithms." ICML 2019. arXiv:1905.12843.
Fairlearn Project. "About Us" (governance, funding, and maintainers) and documentation. https://fairlearn.org/main/about/
Fairlearn Project. "GitHub repository." https://github.com/fairlearn/fairlearn
Fairlearn Organization. "Governance repository." https://github.com/fairlearn/governance
Chouldechova, A. (2017). "Fair Prediction with Disparate Impact." Big Data, 5(2). arXiv:1610.07524.
Kleinberg, J., Mullainathan, S., & Raghavan, M. (2017). "Inherent Trade-Offs in the Fair Determination of Risk Scores." ITCS 2017. arXiv:1609.05807.
Barocas, S., Hardt, M., & Narayanan, A. (2019). Fairness and Machine Learning: Limitations and Opportunities. https://fairmlbook.org/
Bellamy, R. K. E. et al. (2018). "AI Fairness 360." arXiv:1810.01943.
Saleiro, P. et al. (2018). "Aequitas: A Bias and Fairness Audit Toolkit." arXiv:1811.05577.
TensorFlow Team. "Fairness Indicators." https://www.tensorflow.org/responsible_ai/fairness_indicators/guide
Friedler, S. A. et al. (2019). "A Comparative Study of Fairness-Enhancing Interventions in Machine Learning." FAT 2019*.
Holstein, K., Wortman Vaughan, J., Daumé III, H., Dudík, M., & Wallach, H. (2019). "Improving Fairness in Machine Learning Systems: What Do Industry Practitioners Need?" CHI 2019.
Dwork, C., Hardt, M., Pitassi, T., Reingold, O., & Zemel, R. (2012). "Fairness Through Awareness." ITCS 2012.
Kusner, M. J., Loftus, J. R., Russell, C., & Silva, R. (2017). "Counterfactual Fairness." NeurIPS 2017.
Ding, F., Hardt, M., Miller, J., & Schmidt, L. (2021). "Retiring Adult: New Datasets for Fair Machine Learning." NeurIPS 2021.
Lee, M. S. A., & Singh, J. (2021). "The Landscape and Gaps in Open Source Fairness Toolkits." CHI 2021.
Hashimoto, T. B., Srivastava, M., Namkoong, H., & Liang, P. (2018). "Fairness without Demographics in Repeated Loss Minimization." ICML 2018.
Zhang, B. H., Lemoine, B., & Mitchell, M. (2018). "Mitigating Unwanted Biases with Adversarial Learning." AAAI/ACM AIES 2018.
Microsoft. "Responsible AI Toolbox." https://github.com/microsoft/responsible-ai-toolbox
Yang, J. et al. (2024). "OxonFair: A Flexible Toolkit for Algorithmic Fairness." arXiv:2407.13710.
Zemel, R., Wu, Y., Swersky, K., Pitassi, T., & Dwork, C. (2013). "Learning Fair Representations." Proceedings of the 30th International Conference on Machine Learning (ICML), PMLR 28(3):325-333. https://proceedings.mlr.press/v28/zemel13.html
PyPI Stats. "fairlearn download statistics." https://pypistats.org/packages/fairlearn
Fairlearn Project. "Version 0.14.0 release notes" and GitHub Releases. https://fairlearn.org/main/user_guide/installation_and_version_guide/v0.14.0.html

Editing: Fairlearn