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.

翻译：对于安全相关的应用，生成可信的深度神经网络至关重要，其预测需关联能够代表正确概率的置信度以支持后续决策。现有密集二分类模型容易过度自信。为改善模型校准，我们提出自适应随机标签扰动方法，该方法为每张训练图像学习独特的标签扰动水平。该算法采用我们提出的自校准二元交叉熵损失函数，该函数统一了包括随机方法（如DisturbLabel）和标签平滑在内的标签扰动过程，以在校正校准的同时维持分类性能。该算法遵循经典统计力学的最大熵推断原理，针对缺失信息最大化预测熵。其执行过程满足以下条件之一：（1）在保留已知数据分类精度的同时作为保守解，或（2）通过最小化预测精度与目标训练标签期望置信度之间的差距，专门提升模型校准程度。大量实验结果表明，该算法能显著改善密集二分类模型在分布内和分布外数据上的校准程度。代码已开源在https://github.com/Carlisle-Liu/ASLP。