AURC
A classical evaluation metric in the field of selective prediction that measures the area under the curve of risk as a function of coverage when a model ranks predictions by confidence and abstains, treating all errors as equivalent.
Timeline (last 90 days)
eCoMAE reduced the OOD AURC (area under the risk-coverage curve) from 0.0718 to 0.0648.
AURC (Area Under Risk-Coverage Curve) is a metric for measuring the quality of selective prediction; a smaller area under the curve indicates that the model is better able to rank high-risk samples into the rejection region.
Theoretical analysis shows that UtilityAUC can distinguish between models that perform similarly on standard accuracy and AURC.
AURC focuses on error frequency but treats all errors as equivalent, whereas UtilityAUC introduces utility weights so that high-cost errors account for a larger share in the curve integral.
All Facts (4)
eCoMAE reduced the OOD AURC (area under the risk-coverage curve) from 0.0718 to 0.0648.
50%UnverifiedAURC (Area Under Risk-Coverage Curve) is a metric for measuring the quality of selective prediction; a smaller area under the curve indicates that the model is better able to rank high-risk samples into the rejection region.
50%UnverifiedTheoretical analysis shows that UtilityAUC can distinguish between models that perform similarly on standard accuracy and AURC.
50%UnverifiedAURC focuses on error frequency but treats all errors as equivalent, whereas UtilityAUC introduces utility weights so that high-cost errors account for a larger share in the curve integral.
50%