Calibration Error Estimation Using Fuzzy Binning
This work addresses the need for more reliable deep learning frameworks by improving calibration error estimation, though it is incremental as it builds on existing binning methods.
The authors tackled the problem of neural networks being overconfident by proposing a Fuzzy Calibration Error (FCE) metric that uses fuzzy binning to reduce skew in probability estimates, showing it offers better calibration error estimation, particularly in multi-class settings.
Neural network-based decisions tend to be overconfident, where their raw outcome probabilities do not align with the true decision probabilities. Calibration of neural networks is an essential step towards more reliable deep learning frameworks. Prior metrics of calibration error primarily utilize crisp bin membership-based measures. This exacerbates skew in model probabilities and portrays an incomplete picture of calibration error. In this work, we propose a Fuzzy Calibration Error metric (FCE) that utilizes a fuzzy binning approach to calculate calibration error. This approach alleviates the impact of probability skew and provides a tighter estimate while measuring calibration error. We compare our metric with ECE across different data populations and class memberships. Our results show that FCE offers better calibration error estimation, especially in multi-class settings, alleviating the effects of skew in model confidence scores on calibration error estimation. We make our code and supplementary materials available at: https://github.com/bihani-g/fce