Disparate Impact Remover

Fairness Transparency Reliability

Description

Disparate Impact Remover is a preprocessing technique that transforms feature values in a dataset to reduce statistical dependence between features and protected attributes (like race or gender). The method modifies non-protected features through mathematical transformations that preserve the utility of the data whilst reducing correlations that could lead to discriminatory outcomes. This approach specifically targets the '80% rule' disparate impact threshold by adjusting feature distributions to ensure more equitable treatment across demographic groups in downstream model predictions.

Example Use Cases

Fairness

Transforming features in a credit scoring dataset where variables like 'years of employment' and 'education level' are correlated with race, applying mathematical transformations to reduce these correlations whilst preserving the predictive value for creditworthiness assessment.

Preprocessing a recruitment dataset where features like 'previous job titles' and 'university attended' correlate with gender, modifying these features to ensure the '80% rule' is met whilst maintaining useful information for predicting job performance.

Transparency

Preprocessing financial lending data to provide transparent bias metrics showing the quantified reduction in correlation between protected attributes and creditworthiness features, enabling institutions to demonstrate compliance with the 80% rule and explain their fairness interventions to regulators.

Reliability

Ensuring consistent model performance across demographic groups in healthcare risk assessment by mathematically transforming features to reduce protected attribute correlations, improving reliability of predictions for minority populations who may have been systematically under-served.

Limitations

Feature transformations may reduce model accuracy by removing or distorting important predictive information during the debiasing process.
Only addresses measured protected attributes and cannot eliminate bias that operates through unmeasured proxy variables.
Effectiveness depends on the specific transformation method chosen and may not generalise well to different datasets or domains.
May create artificial feature distributions that don't reflect real-world data patterns, potentially causing issues in model deployment.

Resources

holistic-ai/holisticai

Software Package

Comprehensive open-source toolkit for AI fairness with bias measurement, mitigation techniques, and visualisation tools

Disparate Impact Remover — holisticai documentation

Tutorial

Comprehensive tutorial covering theoretical background, methodology, and practical implementation of disparate impact removal

Trusted-AI/AIF360

Software Package

IBM Research's extensible open-source library for detecting and mitigating algorithmic bias across multiple domains

aif360.algorithms.preprocessing.DisparateImpactRemover — aif360 ...

Documentation

Technical API documentation for AIF360's DisparateImpactRemover class with parameters, methods, and usage examples

Related Techniques

Name	Description	Assurance Goals
Model Pruning	Model pruning systematically removes less important weights, neurons, or entire layers from neural networks to create smaller, more efficient models whilst maintaining performance. This process involves iterative removal based on importance criteria (weight magnitudes, gradient information, activation patterns) followed by fine-tuning. Pruning can be structured (removing entire neurons/channels) or unstructured (removing individual weights), with structured pruning providing greater computational benefits and interpretability through simplified architectures.	Explainability Reliability Safety
Red Teaming	Red teaming involves systematic adversarial testing of AI/ML systems by dedicated specialists who attempt to identify flaws, vulnerabilities, harmful outputs, and ways to circumvent safety measures. Drawing from cybersecurity practices, red teams employ diverse attack vectors including prompt injection, adversarial examples, edge case exploitation, social engineering scenarios, and goal misalignment probes. Unlike standard testing that validates expected behaviour, red teaming specifically seeks to break systems through creative and adversarial approaches, revealing non-obvious risks and failure modes that could be exploited maliciously or cause harm in deployment.	Safety Reliability Fairness Security
Threshold Optimiser	Threshold Optimiser adjusts decision thresholds for different demographic groups after model training to satisfy specific fairness constraints. This post-processing technique optimises group-specific thresholds by analysing the probability distribution of model outputs, allowing practitioners to achieve fairness goals like demographic parity or equalised opportunity without modifying the underlying model. The optimiser finds optimal threshold values for each group that balance fairness requirements with overall model performance, making it particularly useful when fairness considerations arise after model deployment.	Fairness
Data Version Control	Data Version Control (DVC) is a Git-like version control system specifically designed for machine learning data, models, and experiments. It tracks changes to large data files, maintains reproducible ML pipelines, and creates a complete audit trail of data transformations, model training, and evaluation processes. DVC works alongside Git to provide end-to-end lineage tracking from raw data through preprocessing, training, and deployment, enabling teams to reproduce any model version and understand exactly how datasets evolved throughout the ML lifecycle.	Transparency Reliability
Monotonicity Constraints	Monotonicity constraints enforce consistent directional relationships between input features and model predictions, ensuring that increasing a feature value either always increases, always decreases, or has no effect on the output. These constraints integrate domain knowledge into model training, preventing counterintuitive relationships that may arise from spurious correlations in data. By maintaining logical feature relationships (e.g., experience always positively influences salary), monotonicity constraints enhance model trustworthiness, interpretability, and alignment with business logic whilst often improving generalisation to new data.	Transparency Reliability
Anomaly Detection	Anomaly detection identifies unusual behaviours, inputs, or outputs that deviate significantly from established normal patterns using statistical, machine learning, or rule-based methods. Applied to AI/ML systems, it serves as a continuous monitoring mechanism that can flag unexpected model predictions, suspicious input patterns, data drift, adversarial attacks, or operational malfunctions. By establishing baselines of normal system behaviour and alerting when deviations exceed predefined thresholds, organisations can detect potential security threats, model degradation, fairness violations, or system failures before they cause significant harm.	Safety Reliability Fairness Security

Tags

Applicable Models:

Data Requirements:

Sensitive Attributes

Data Type:

Evidence Type:

Quantitative Metric

Expertise Needed:

Fairness Approach:

Lifecycle Stage:

Model Development

Technique Type: