Despite the great success of state-of-the-art deep neural networks, several studies have reported models to be over-confident in predictions, indicating miscalibration. Label Smoothing has been proposed as a solution to the over-confidence problem and works by softening hard targets during training, typically by distributing part of the probability mass from a `one-hot' label uniformly to all other labels. However, neither model nor human confidence in a label are likely to be uniformly distributed in this manner, with some labels more likely to be confused than others. In this paper we integrate notions of model confidence and human confidence with label smoothing, respectively \textit{Model Confidence LS} and \textit{Human Confidence LS}, to achieve better model calibration and generalization. To enhance model generalization, we show how our model and human confidence scores can be successfully applied to curriculum learning, a training strategy inspired by learning of `easier to harder' tasks. A higher model or human confidence score indicates a more recognisable and therefore easier sample, and can therefore be used as a scoring function to rank samples in curriculum learning. We evaluate our proposed methods with four state-of-the-art architectures for image and text classification task, using datasets with multi-rater label annotations by humans. We report that integrating model or human confidence information in label smoothing and curriculum learning improves both model performance and model calibration. The code are available at \url{https://github.com/AoShuang92/Confidence_Calibration_CL}.
翻译:尽管最先进的深层神经网络取得了巨大成功,但有几项研究报告说模型在预测中过于自信,这表明了错误校正。在本文中,我们把模型信心和人类信心的概念与标签平滑(分别是\textit{Model Infority LS}和\textit{Human Inful LS})结合起来,目的是在培训期间实现更好的模型校准和概括化。为了加强模型的概括化,我们展示了我们的模型和人类信心分数如何能够成功地用于课程学习,而根据学习“更易的模型”而激发的培训战略。一个更高的模型或人类信用分数表示更加可辨性,因此比较容易的样本,因此,我们可以在结构平滑的标签中,分别纳入\ textitle A{Model InfenceLS} 和\ textitleitalitalital{人类信心LS},以便实现更好的模型校正校正校准。我们提出的模型和图解方法是,通过学习人类的模型和图象学课程,我们建议用模型和图象学的图解方法来进行人类学习。