An Entropic Metric for Measuring Calibration of Machine Learning Models
This work addresses the need for better calibration metrics in machine learning, particularly for binary classification, though it is incremental as it adapts concepts from target tracking.
The paper tackles the problem of measuring calibration in machine learning models by proposing a new metric, the Entropic Calibration Difference (ECD), which distinguishes between under- and over-confidence, and demonstrates its performance on real and simulated data compared to existing metrics like ECE and ESCE.
Understanding the confidence with which a machine learning model classifies an input datum is an important, and perhaps under-investigated, concept. In this paper, we propose a new calibration metric, the Entropic Calibration Difference (ECD). Based on existing research in the field of state estimation, specifically target tracking (TT), we show how ECD may be applied to binary classification machine learning models. We describe the relative importance of under- and over-confidence and how they are not conflated in the TT literature. Indeed, our metric distinguishes under- from over-confidence. We consider this important given that algorithms that are under-confident are likely to be 'safer' than algorithms that are over-confident, albeit at the expense of also being over-cautious and so statistically inefficient. We demonstrate how this new metric performs on real and simulated data and compare with other metrics for machine learning model probability calibration, including the Expected Calibration Error (ECE) and its signed counterpart, the Expected Signed Calibration Error (ESCE).