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.
Online learning with an unknown fairness metric
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.