search

shivashankarrs / classimb_fairness

5 stars 0 forks source link

Task 5 (debiasing approaches) #5

Open shivashankarrs opened 3 years ago

shivashankarrs commented 3 years ago

To benchmark our results for fairness, we can compare with debiasing approaches.

shivashankarrs commented 3 years ago

INLP and adversarial.

https://github.com/shauli-ravfogel/nullspace_projection

https://aif360.readthedocs.io/en/v0.2.3/modules/inprocessing.html#adversarial-debiasing