AIF360
AIF360 copied to clipboard
Trusted-AI
A comprehensive set of fairness metrics for datasets and machine learning models, explanations for these metrics, and algorithms to mitigate bias in datasets and models.
The widely used german credit data (that is already available in the toolkit) apparently has coding errors, so consider including https://archive.ics.uci.edu/ml/datasets/South+German+Credit+%28UPDATE%29 http://www1.beuth-hochschule.de/FB_II/reports/Report-2019-004.pdf
Added the "Introduction to AI Fairness 360" vignette to the R package. Linked to issue #194
- Added `predict_proba` for `RejectOptionClassification` in the sklearn-compatible version of AIF360. - Adjusted `GridSearch` and `ExponentiatedGradient` classes to be compatible with Fairlearn 0.7.0. Worked on as part of an internship...
I'm trying to use the ExponentiatedGradientReduction-function for my classifier with the following code: ``` from aif360.algorithms.inprocessing import ExponentiatedGradientReduction as EGR from sklearn.linear_model import LogisticRegression as logreg clf = logreg(random_state=1) EGR(estimator=clf,...
Major improvements: - Changed how `prot_attr` arguments are handled. Now, when processing a dataset and running metrics, an explicit array (or list of arrays) containing protected attribute values per sample...
Hi there! In the paper, the authors mentioned the strategy to reach demographic parity or equality of odds by changing slightly the architecture of the network. How can this option...
I'm trying to run the disparate impact remover on a dataset - however after varying the repair value between 0 -> 1.0 I see no change in disparate impact metric....
Two of the Classification Metric tests fail in the AIF360 R pkg: - between_all_groups_coefficient_of_variation() - coefficient_of_variation()
The Learning Fair Representations class (LFR) parameters are `LFR(unprivileged_groups, privileged_groups, k=5, Ax=0.01, Ay=1.0, Az=50.0, print_interval=250, verbose=0, seed=None)`. The only required parameters are unprivileged_groups and privileged_groups, so I have used `LFR...
Metadata
Owner
Metadata
A comprehensive set of fairness metrics for datasets and machine learning models, explanations for these metrics, and algorithms to mitigate bias in datasets and models.