Unintended Bias and Fairness

Protected attributes are referred to as features that may not be used as the basis for decisions (for example, race, gender, etc.). When machine learning is applied to decision-making processes involving humans, one should not only look for models with good performance, but also models that do not discriminate against protected population subgroups.

Oracle Guardian AI Project provides metrics dedicated to assessing and measuring the compliance of a model or a dataset with a fairness metric. The provided metrics all correspond to different notions of fairness, which the user should carefully select from while taking into account their problem’s specificities.

It also provides a bias mitigation algorithm that fine-tunes decison thresholds across demographic groups to compensate for the bias present in the original model. The approach is called Bias Mitigation.

• Fairness Metrics
  • Overview of the Fairness Metrics
  • Unintended Bias Detection
    • Measure the Compliance of a Model with a Fairness Metric
    • Measure the Compliance of the True Labels of a Dataset with a Fairness Metric
    • Other Fairness Metrics
• Bias Mitigation