Calibration and fairness

In this study, I analyzed the interplay between calibration and fairness measures and found that it is possible tob achieve fairness while maintaining calibration through proper thresholding of the non-binary scoresnassociated with classification tasks. My findings highlight the complexity of balancing calibrated performance with fairness considerations in machine learning applications and the importance of carefully selecting the threshold in order to achieve the desired fairness metric. Here's is a link to my full paper!