Function reference • mlr3fairness

Measuring Bias

Performance measures for fairness, based on [Measure].

mlr_measures_fairness: Fairness Measures in mlr3

MeasureFairness: Base Measure for Fairness

MeasureFairnessComposite: Composite Fairness Measure

MeasureFairnessConstraint: Fairness Constraint Measure

MeasureSubgroup: Evaluate a metric on a subgroup

groupdiff_tau() groupdiff_absdiff() groupdiff_diff(): Groupwise Operations

groupwise_metrics(): Evaluate a metric on each protected subgroup in a task.

compute_metrics(): Compute metrics for non-mlr3 predictions.

mlr_measures_positive_probability MeasurePositiveProbability: Positive Probability Measure

Detecting Bias

fairness_tensor(): Compute the Fairness Tensor given a Prediction and a Task

fairness_accuracy_tradeoff(): Plot Fairness Accuracy Trade-offs

fairness_prediction_density(): Probability Density Plot

compare_metrics(): Compare different metrics

Correcting Bias

mlr_pipeops_equalized_odds PipeOpEOd: Equalized Odds Debiasing

mlr_pipeops_reweighing PipeOpReweighingWeights PipeOpReweighingOversampling: Reweighing to balance disparate impact metric

mlr_pipeops_explicit_pta PipeOpExplicitPta: PipeOpExplicitPta

Fair Learners

mlr_learners_fairness: Fair Learners in mlr3

Reports

report_datasheet(): Create a Datasheet for Documenting a Dataset

report_modelcard(): Create a Modelcard

report_fairness(): Create a Fairness Report

Integrated data & tasks

adult adult_test adult_train: Adult Dataset

compas Compas: COMPAS Dataset

task_summary(): Task summary for fairness report