Stop Measuring Calibration When Humans Disagree
Joris Baan, Wilker Ferreira Aziz, Barbara Plank, Raquel Fernández · 2022
Calibration is a popular framework to evaluate whether a classifier knows when it does not know-i.e., its predictive probabilities are a good indication of how likely a prediction is to be correct.Correctness is commonly estimated against the human majority class.Recently, calibration to human majority has been measured on tasks where humans inherently disagree about which class applies.We show that measuring calibration to human majority given inherent disagreements is theoretically problematic, demonstrate this empirically on the ChaosNLI dataset, and derive several instancelevel measures of calibration that capture key statistical properties of human judgementsclass frequency, ranking and entropy. 1