Model Calibration in Dense Classification with Adaptive Label Perturbation

TL;DR

This paper introduces Adaptive Stochastic Label Perturbation (ASLP), a novel method to improve the calibration of dense binary classification models by learning individual label perturbations and unifying calibration techniques.

Contribution

The paper proposes ASLP, a new calibration method that learns per-image label perturbations and combines stochastic approaches with label smoothing for better trustworthiness.

Findings

ASLP significantly improves model calibration on in-distribution data.

ASLP enhances calibration on out-of-distribution data.

The method maintains classification accuracy while improving confidence estimates.

Abstract

For safety-related applications, it is crucial to produce trustworthy deep neural networks whose prediction is associated with confidence that can represent the likelihood of correctness for subsequent decision-making. Existing dense binary classification models are prone to being over-confident. To improve model calibration, we propose Adaptive Stochastic Label Perturbation (ASLP) which learns a unique label perturbation level for each training image. ASLP employs our proposed Self-Calibrating Binary Cross Entropy (SC-BCE) loss, which unifies label perturbation processes including stochastic approaches (like DisturbLabel), and label smoothing, to correct calibration while maintaining classification rates. ASLP follows Maximum Entropy Inference of classic statistical mechanics to maximise prediction entropy with respect to missing information. It performs this while: (1) preserving classification accuracy on known data as a conservative solution, or (2) specifically improves model calibration degree by minimising the gap between the prediction accuracy and expected confidence of the target training label. Extensive results demonstrate that ASLP can significantly improve calibration degrees of dense binary classification models on both in-distribution and out-of-distribution data. The code is available on https://github.com/Carlisle-Liu/ASLP.