论文标题
域漂移方案的事后不确定性校准
Post-hoc Uncertainty Calibration for Domain Drift Scenarios
论文作者
论文摘要
我们解决了不确定性校准的问题。尽管标准的深神经网络通常会产生未校准的预测，但可以使用事后校准方法来实现预测的真实可能性的校准置信得分。但是，迄今为止，这些方法的重点一直放在内域校准上。我们的贡献是两个方面。首先，我们表明现有的事后校准方法在域移位下产生了高度过度自信的预测。其次，我们引入了一个简单的策略，在执行事后校准步骤之前，将扰动应用于验证集中的样品。在广泛的实验中，我们证明了这种扰动步骤在域转移下在各种架构和建模任务上的校准下实现了更好的校准。
We address the problem of uncertainty calibration. While standard deep neural networks typically yield uncalibrated predictions, calibrated confidence scores that are representative of the true likelihood of a prediction can be achieved using post-hoc calibration methods. However, to date the focus of these approaches has been on in-domain calibration. Our contribution is two-fold. First, we show that existing post-hoc calibration methods yield highly over-confident predictions under domain shift. Second, we introduce a simple strategy where perturbations are applied to samples in the validation set before performing the post-hoc calibration step. In extensive experiments, we demonstrate that this perturbation step results in substantially better calibration under domain shift on a wide range of architectures and modelling tasks.
