Manojit Nandi

ENH Updated EY Notebook

13

This is an in-progress version of the [EY whitepaper](https://www.microsoft.com/en-us/research/uploads/prod/2020/09/Fairlearn-EY_WhitePaper-2020-09-22.pdf) notebook. The main goal of this notebook is to highlight some of the functionality of Fairlearn (e.g. how to select an...

ENH Add "Doubly-Corrected" Variance Estimator from "De-biasing 'bias' measurement" by Lum, Zhang, and Bower

8

In the paper [De-biasing "bias" measurements](https://arxiv.org/abs/2205.05770), the authors introduce a "corrected" estimator for the between-group variance. The authors state that commonly used fairness estimators consistently over-estimate the "true level" of...

#### Describe the issue linked to the documentation We want to update sections *1.1* and *1.2* of the User Guide to be more comprehensive, based on the [Overview of Fairness](https://github.com/fairlearn/talks/blob/main/2021_scipy_tutorial/overview.pdf)...

Quick fix to the Student version and the teacher version of the tutorial to fix the definition of *False Positive Rate* in one of the exercises. Signed-off-by: Manojit Nandi