Given we detected some form of bias during bias auditing, we are often interested in obtaining fair(er) models. There are several ways to achieve this, such as collecting additional data or finding and fixing errors in the data, but given there are no biases in the labelling process one other option is to debias models using either preprocessing, postprocessing and inprocessing methods.
mlr3fairness provides some operators as
mlr3pipelines. If you are not familiar with mlr3pipelines, the mlr3 book
We again showcase debiasing using the
mlr3fairness implements 2 reweighing-based algorithms:
reweighing_wts adds observation weights to a
Task that can counteract imbalances between the conditional probabilities \(P(Y | pta)\).
We fist instantiate the
p1 = po("reweighing_wts")
and directly add the weights:
t1 = p1$train(list(task))[]
Often we directly combine the
PipeOp with a
Learner to automate the preprocessing (see
learner_rw). Below we instantiate a small benchmark
set.seed(4321) learner = lrn("classif.rpart", cp = 0.005) learner_rw = as_learner(po("reweighing_wts") %>>% learner) grd = benchmark_grid(list(task), list(learner, learner_rw), rsmp("cv", folds=3)) bmr = benchmark(grd)
We can now compute the metrics for our benchmark and see if reweighing actually improved fairness, measured via True Positive Rate (TPR) and classification accuracy (ACC):
bmr$aggregate(msrs(c("fairness.tpr", "fairness.acc"))) #> nr resample_result task_id learner_id #> 1: 1 <ResampleResult> adult_train classif.rpart #> 2: 2 <ResampleResult> adult_train reweighing_wts.classif.rpart #> resampling_id iters fairness.tpr fairness.acc #> 1: cv 3 0.07494903 0.1162688 #> 2: cv 3 0.01151982 0.1054431
Our model became way fairer wrt. TPR but minimally worse wrt. accuracy!