ember

7Calibration

A reliability diagram plots each confidence bin's accuracy against its mean confidence; a perfectly calibrated advisor sits on the diagonal [11]. Bars above the diagonal mean ember was more often right than its confidence claimed (under-confidence); bars below it mean over-confidence, the dangerous direction for threshold-driven agents.

Reliability of noul answersAccuracy bars per confidence bin against the diagonal of perfect calibration.accuracymean confidenceperfect calibration0.00.00.20.20.40.40.60.60.80.81.01.0Top-label confidenceAccuracyn=55n=57n=69n=105n=438ECE 0.065 MCE 0.170
Figure 6 Reliability of noul answers (724 questions). Coral lines mark each bin's mean confidence; shading below a line marks over-confidence.
TypeNMean confidenceAccuracyAccuracy minus confidenceECEMCEBrier
noul7240.87180.5%-0.065 (over-confident)0.0650.1700.133