Extending confidence calibration to generalised measures of variation
This work addresses the need for more general calibration metrics in machine learning, but it is incremental as it builds upon the established ECE framework.
The authors tackled the problem of assessing calibration in machine learning classifiers by proposing the Variation Calibration Error (VCE) metric, which extends the Expected Calibration Error (ECE) to handle any metric of variation, such as Shannon entropy, and demonstrated that VCE approaches zero with increasing data samples on perfectly calibrated synthetic predictions, unlike the existing UCE metric.
We propose the Variation Calibration Error (VCE) metric for assessing the calibration of machine learning classifiers. The metric can be viewed as an extension of the well-known Expected Calibration Error (ECE) which assesses the calibration of the maximum probability or confidence. Other ways of measuring the variation of a probability distribution exist which have the advantage of taking into account the full probability distribution, for example the Shannon entropy. We show how the ECE approach can be extended from assessing confidence calibration to assessing the calibration of any metric of variation. We present numerical examples upon synthetic predictions which are perfectly calibrated by design, demonstrating that, in this scenario, the VCE has the desired property of approaching zero as the number of data samples increases, in contrast to another entropy-based calibration metric (the UCE) which has been proposed in the literature.