Using the FiND world, a hypothetical fair world where protected attributes have no causal effect on the target, the authors show that fairness metrics become compatible and fairness aligns with accuracy, and that pre-processing methods can approximate this world in practice.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Overcoming Fairness Trade-offs via Pre-processing: A Causal Perspective
Using the FiND world, a hypothetical fair world where protected attributes have no causal effect on the target, the authors show that fairness metrics become compatible and fairness aligns with accuracy, and that pre-processing methods can approximate this world in practice.