fairgbm
fairgbm copied to clipboard
Implement constraints on predicted prevalence (demographic parity)
Summary
We should enable creating group-wise constraints on the percentage of positive predictions (a.k.a., predicted prevalence).
This enables the popular Demographic Parity fairness metric.
TODO
- [ ] Add
constraint_type=PPto the available group-wise constraint options; - [ ] Implement gradient descent on proxy predicted prevalence metrics;
- [ ] Add the same
PPconstraint type for global constraints- This would be a constraint on maximum alert rate -> quite useful for our in-house use cases
EDIT
Minor notes for the future PP constraint implementation: 83de2426-ea6e-4308-9f0d-5ddc3b001580.pdf