A meta-learning model with a fairness regularizer can be fine-tuned from a handful of labeled examples, and an interpretable surrogate model can warn when mean shifts would break group fairness.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Fairness Warnings and Fair-MAML: Learning Fairly with Minimal Data
A meta-learning model with a fairness regularizer can be fine-tuned from a handful of labeled examples, and an interpretable surrogate model can warn when mean shifts would break group fairness.