data requirements
sensitive attributes
Needs data labeled with protected attributes
26 techniques
| Goals | Models | Data Types | Description | |||
|---|---|---|---|---|---|---|
| Demographic Parity Assessment | Algorithmic | Model Agnostic | Any | Demographic Parity Assessment evaluates whether a model produces equal positive prediction rates across different... | ||
| Adversarial Debiasing | Algorithmic | Neural Network | Any | Adversarial debiasing reduces bias by training models using a competitive adversarial setup, similar to Generative... | ||
| Counterfactual Fairness Assessment | Algorithmic | Model Agnostic | Any | Counterfactual Fairness Assessment evaluates whether a model's predictions would remain unchanged if an individual's... | ||
| Sensitivity Analysis for Fairness | Algorithmic | Model Agnostic | Any | Sensitivity Analysis for Fairness systematically evaluates how model predictions change when sensitive attributes or... | ||
| Model Cards | Documentation | Model Agnostic | Any | Model cards are standardised documentation frameworks that systematically document machine learning models through... | ||
| Reweighing | Algorithmic | Model Agnostic | Any | Reweighing is a pre-processing technique that mitigates bias by assigning different weights to training examples based... | ||
| Disparate Impact Remover | Algorithmic | Model Agnostic | Tabular | Disparate Impact Remover is a preprocessing technique that transforms feature values in a dataset to reduce statistical... | ||
| Fairness GAN | Algorithmic | GAN | Any | A data generation technique that employs Generative Adversarial Networks (GANs) to create fair synthetic datasets by... | ||
| Relabelling | Procedural | Model Agnostic | Any | A preprocessing fairness technique that modifies class labels in training data to achieve equal positive outcome rates... | ||
| Preferential Sampling | Procedural | Model Agnostic | Any | A preprocessing fairness technique developed by Kamiran and Calders that addresses dataset imbalances by re-sampling... | ||
| Attribute Removal (Fairness Through Unawareness) | Algorithmic | Model Agnostic | Any | Attribute Removal (Fairness Through Unawareness) ensures fairness by completely excluding protected attributes such as... | ||
| Fair Adversarial Networks | Algorithmic | Neural Network CNN +1 | Any | An in-processing fairness technique that employs adversarial training with dual neural networks to learn fair... | ||
| Prejudice Remover Regulariser | Algorithmic | Logistic Regression Probabilistic | Tabular | An in-processing fairness technique that adds a fairness penalty to machine learning models to reduce bias against... | ||
| Meta Fair Classifier | Algorithmic | Model Agnostic | Any | An in-processing fairness technique that employs meta-learning to modify any existing classifier for optimising fairness... | ||
| Exponentiated Gradient Reduction | Algorithmic | Model Agnostic | Any | An in-processing fairness technique based on Agarwal et al.'s reductions approach that transforms fair classification... | ||
| Fair Transfer Learning | Algorithmic | Model Agnostic | Any | An in-processing fairness technique that adapts pre-trained models from one domain to another whilst explicitly... | ||
| Adaptive Sensitive Reweighting | Algorithmic | Model Agnostic | Any | Adaptive Sensitive Reweighting dynamically adjusts the importance of training examples during model training based on... | ||
| Multi-Accuracy Boosting | Algorithmic | Model Agnostic Ensemble | Any | An in-processing fairness technique that employs boosting algorithms to improve accuracy uniformly across demographic... | ||
| Equalised Odds Post-Processing | Algorithmic | Model Agnostic | Any | A post-processing fairness technique based on Hardt et al.'s seminal work that adjusts classification thresholds after... | ||
| Threshold Optimiser | Algorithmic | Model Agnostic | Any | Threshold Optimiser adjusts decision thresholds for different demographic groups after model training to satisfy... |
Rows per page
Page 1 of 2