Privilege scores quantify the difference between real-world model predictions and predictions in a fair world with the protected attribute's influence removed, plus Shapley-based contributions explaining each privilege path.
Causal Fair Machine Learning via Rank - Preserving Interventional Distributions
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Privilege Scores
Privilege scores quantify the difference between real-world model predictions and predictions in a fair world with the protected attribute's influence removed, plus Shapley-based contributions explaining each privilege path.