G-FORCE runs separate multiplicative-weights instances for each protected group and label, and tunes the mixing probabilities to approximately equalize false positive and false negative rates across groups.
Empirical risk minimization under fairness constraints
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Towards Reducing Biases in Combining Multiple Experts Online
G-FORCE runs separate multiplicative-weights instances for each protected group and label, and tunes the mixing probabilities to approximately equalize false positive and false negative rates across groups.