Multiclass Alignment of Confidence and Certainty for Network Calibration

TL;DR

This paper introduces MACC, a simple train-time auxiliary loss that improves neural network calibration by aligning confidence and certainty, leading to state-of-the-art results across diverse datasets.

Contribution

The paper proposes MACC, a novel plug-and-play auxiliary loss for train-time calibration that explicitly aligns confidence and certainty to enhance model calibration.

Findings

Achieves state-of-the-art calibration on ten datasets.

Effective for in-domain and out-of-domain predictions.

Applicable across various recognition and medical imaging tasks.

Abstract

Deep neural networks (DNNs) have made great strides in pushing the state-of-the-art in several challenging domains. Recent studies reveal that they are prone to making overconfident predictions. This greatly reduces the overall trust in model predictions, especially in safety-critical applications. Early work in improving model calibration employs post-processing techniques which rely on limited parameters and require a hold-out set. Some recent train-time calibration methods, which involve all model parameters, can outperform the postprocessing methods. To this end, we propose a new train-time calibration method, which features a simple, plug-and-play auxiliary loss known as multi-class alignment of predictive mean confidence and predictive certainty (MACC). It is based on the observation that a model miscalibration is directly related to its predictive certainty, so a higher gap between the mean confidence and certainty amounts to a poor calibration both for in-distribution and out-of-distribution predictions. Armed with this insight, our proposed loss explicitly encourages a confident (or underconfident) model to also provide a low (or high) spread in the presoftmax distribution. Extensive experiments on ten challenging datasets, covering in-domain, out-domain, non-visual recognition and medical image classification scenarios, show that our method achieves state-of-the-art calibration performance for both in-domain and out-domain predictions. Our code and models will be publicly released.