Concept机会均等 / 均等机会

Equalized Odds

A fairness metric in algorithmic decision-making that requires a model to maintain equal true positive rates and false positive rates across different demographic groups, ensuring that the model's error probabilities do not vary systematically with protected attributes.

Timeline (last 90 days)

Oct 4

Fairness is measured by Equalized Odds, a metric that requires the model to maintain consistent true positive rates and false positive rates across different demographic groups.

Unverified50%

All Facts (1)

Unverified

Fairness is measured by Equalized Odds, a metric that requires the model to maintain consistent true positive rates and false positive rates across different demographic groups.

50%

Source Articles

Fairness Meets Explainability: The Challenges of a Reinforcement Learning Multi-Instance Student Early Warning SystemOct 4