ρ-Norm Scaling, a temperature-scaling variant with a learnable norm and KL regularization, reduces expected calibration error on several image classifiers without changing accuracy.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Parametric $\rho$-Norm Scaling Calibration
ρ-Norm Scaling, a temperature-scaling variant with a learnable norm and KL regularization, reduces expected calibration error on several image classifiers without changing accuracy.