Equal Opportunity Difference

Description

A fairness metric that quantifies discrimination by measuring the difference in true positive rates (recall) between protected and privileged groups. Based on Hardt et al.'s equality of opportunity framework, this metric computes the maximum difference in TPR across demographic groups, with a value of 0 indicating perfect fairness. The technique provides a mathematical measure of whether qualified individuals from different groups have equal chances of receiving positive predictions.

Example Use Cases

Fairness

Evaluating a loan approval model to ensure that qualified applicants from different ethnic backgrounds have equal approval rates, measuring whether a 90% TPR for qualified white applicants is matched by similar rates for qualified minority applicants.

Transparency

Auditing a medical diagnosis system to verify that patients with a particular condition are detected at equal rates across age groups, providing transparent evidence that diagnostic accuracy is consistent regardless of patient demographics.

Reliability

Monitoring a university admissions algorithm in production to ensure that qualified students from different socioeconomic backgrounds have equal acceptance rates, validating the reliability of the fairness properties over time.

Limitations

Only considers true positive rates and ignores false positive rate disparities, potentially allowing discrimination in the form of unequal false alarm rates between groups.
Requires accurate ground truth labels for the positive class, which may be biased or unavailable in some domains.
Improving TPR for one group might increase FPR for that group, creating trade-offs between different types of fairness.
Does not account for intersectional fairness across multiple protected attributes simultaneously.
May not capture fairness concerns for the negative class or individuals who should not receive positive predictions.

Resources

aif360.sklearn.metrics.equal_opportunity_difference — aif360 0.6.1 ...

Documentation

Welcome to TransparentAI's documentation! — TransparentAI 0.1.0 ...

Documentation

lale.lib.aif360.util module — LALE 0.9.0-dev documentation

Documentation

IBM/bias-mitigation-of-machine-learning-models-using-aif360

Software Package

aif360.algorithms.postprocessing.RejectOptionClassification ...

Documentation

Related Techniques

Name	Description	Assurance Goals
Exponentiated Gradient Reduction	An in-processing fairness technique based on Agarwal et al.'s reductions approach that transforms fair classification into a sequence of cost-sensitive classification problems. The method uses an exponentiated gradient algorithm to iteratively reweight training data, returning a randomised classifier that achieves the lowest empirical error whilst satisfying fairness constraints. This reduction-based framework provides theoretical guarantees about both accuracy and constraint violation, making it suitable for various fairness criteria including demographic parity and equalised odds.	Fairness Transparency Reliability
Neuron Activation Analysis	Neuron activation analysis examines the firing patterns of individual neurons in neural networks by probing them with diverse inputs and analysing their activation responses. This technique helps understand what concepts, features, or patterns different neurons have learned to recognise, providing insights into the model's internal representations. For large language models, this can reveal neurons specialised for linguistic concepts, semantic categories, or even potentially harmful patterns, enabling targeted interventions and deeper model understanding.	Explainability Safety Fairness
Preferential Sampling	A preprocessing fairness technique developed by Kamiran and Calders that addresses dataset imbalances by re-sampling training data with preference for underrepresented groups to achieve discrimination-free classification. This method modifies the training distribution by prioritising borderline objects (instances near decision boundaries) from underrepresented groups for duplication whilst potentially removing instances from overrepresented groups. Unlike relabelling approaches, preferential sampling maintains original class labels whilst creating a more balanced dataset that prevents models from learning biased patterns due to skewed group representation.	Fairness Reliability Transparency
Adaptive Sensitive Reweighting	Adaptive Sensitive Reweighting dynamically adjusts the importance of training examples during model training based on real-time performance across different demographic groups. Unlike traditional static reweighting that fixes weights at the start, this technique continuously monitors fairness metrics and automatically increases the weight of examples from underperforming groups whilst decreasing weights for overrepresented groups. The adaptive mechanism prevents models from perpetuating historical biases by ensuring balanced learning across all demographics throughout the training process.	Fairness Reliability
Generalized Additive Models	An intrinsically interpretable modelling technique that extends linear models by allowing flexible, nonlinear relationships between individual features and the target whilst maintaining the additive structure that preserves transparency. Each feature's effect is modelled separately as a smooth function, visualised as a curve showing how the feature influences predictions across its range. GAMs achieve this through spline functions or other smoothing techniques that capture complex patterns in individual variables without interactions, making them particularly valuable for domains requiring both predictive accuracy and model interpretability.	Transparency Explainability Reliability
Datasheets for Datasets	Datasheets for datasets establish comprehensive documentation standards for datasets, systematically recording creation methodology, data composition, collection procedures, preprocessing transformations, intended applications, potential biases, privacy considerations, and maintenance protocols. These structured documents enhance dataset transparency by providing essential context for appropriate usage, enabling informed decisions about dataset suitability for specific tasks, supporting bias detection and mitigation efforts, ensuring compliance with data protection regulations, and promoting responsible data stewardship throughout the entire data lifecycle from collection to disposal.	Transparency Fairness Privacy

Equal Opportunity Difference

Description

Example Use Cases

Fairness

Transparency

Reliability

Limitations

Resources

aif360.sklearn.metrics.equal_opportunity_difference — aif360 0.6.1 ...

Welcome to TransparentAI's documentation! — TransparentAI 0.1.0 ...

lale.lib.aif360.util module — LALE 0.9.0-dev documentation

IBM/bias-mitigation-of-machine-learning-models-using-aif360

aif360.algorithms.postprocessing.RejectOptionClassification ...

Related Techniques

Tags