Fair Gated Algorithm for Targeted Equity

---

Documentation for package ‘fairGATE’ version 1.0.0

• DESCRIPTION file.
• User guides, package vignettes and other documentation.
• Package NEWS.

Help Pages

adult_ready_small	Adult ready (small sample)
analyse_experts	Analyse and Visualise Expert Network Specialisation
analyse_gnn_results	Analyse and Visualise GNN Results
export_f360_csv	Export predictions for IBM Fairness 360
fairgate	Fit fairGATE in One Call
fairgate_base_grid	Default Hyperparameter Grids for the Base Learner
fairgate_base_learner	Cross-Fitted Base Learner (XGBoost or Neural Network)
fairgate_crossed_penalty	Crossed Variance-Component Penalty for Intersectional Groups
fairgate_crossed_plan	Cell Structure for Crossed (Intersectional) Variance Components
fairgate_disparity_target	Per-Person Disparity-Contribution Target
fairgate_draw_configs	Draw fairGATE Configurations
fairgate_export	Per-Person Export of a fairGATE Fit
fairgate_fair_penalty	fairGATE Fairness Penalty from Individual Predictions
fairgate_gate_map	Gate Maps: softmax, sparsemax and entmax-1.5
fairgate_group_disparity	Multi-Group Separation Disparity: Per Group, Worst Group and Joint
fairgate_group_gap_penalty	Worst-Group Conditional Mean Gap Penalty (continuous outcomes)
fairgate_lambda_ladder	Frontier Ladder of Expert Fairness Weights
fairgate_penalty_from_cells	Between-Group Variance Components for the fairGATE Fairness Penalty
fairgate_penalty_plan	Cell Structure for the fairGATE Fairness Penalty
fairgate_space	fairGATE Search Space and Fixed Settings
fairgate_threshold_penalty	Differentiable Penalty on Thresholded Group Error Rates
fairgate_threshold_plan	Cell Structure for the Thresholded Error-Rate Penalty
fairgate_tune_base_learner	Tune the Base Learner on an Inner Split
plot_sankey	Create a Sankey Plot (robust; aggregates to one row per subject)
prepare_data	Prepare Data for GNN Training
train_gnn	Train and Evaluate the Gated Neural Network (robust splits + safe ROC)