协变量偏移的检测时间批统计校准
Test-time Batch Statistics Calibration for Covariate Shift
摘要：
由于协变量偏移，深度神经网络在应用于看不见的环境时具有明显的退化。像领域自适应这样的传统方法需要预先收集目标数据进行迭代训练，这在现实世界的应用中是不切实际的。在本文中，我们建议在推理过程中使深度模型适应新的环境。以前的解决方案是测试时间标准化，它用目标批处理统计数据代替BN层中的源统计数据。然而，我们发现，测试时间归一化可能会恶化的歧视性结构，由于目标批次统计和源参数之间的不匹配。为此，我们提出了一个通用的配方$α$ -BN校准的批量统计混合的源和目标统计，以减轻域移位和保留的歧视性结构。在$α$ -BN的基础上，进一步提出了一种新的损失函数，形成了一个统一的测试时自适应框架Core，实现了两两类相关性的在线优化。大量的实验表明，我们的方法达到了国家的最先进的性能，共12个数据集，从三个主题，包括模型的鲁棒性损坏，领域泛化的图像分类和语义分割。特别是，我们的$α$在没有任何训练的情况下，在GTA 5 $\rightarrow$ Cityscapes上提高了28.4%到43.9%，甚至优于最新的无源领域自适应方法。
英文摘要：
Deep neural networks have a clear degradation when applying to the unseen environment due to the covariate shift. Conventional approaches like domain adaptation requires the pre-collected target data for iterative training, which is impractical in real-world applications. In this paper, we propose to adapt the deep models to the novel environment during inference. An previous solution is test time normalization, which substitutes the source statistics in BN layers with the target batch statistics. However, we show that test time normalization may potentially deteriorate the discriminative structures due to the mismatch between target batch statistics and source parameters. To this end, we present a general formulation $α$-BN to calibrate the batch statistics by mixing up the source and target statistics for both alleviating the domain shift and preserving the discriminative structures. Based on $α$-BN, we further present a novel loss function to form a unified test time adaptation framework Core, which performs the pairwise class correlation online optimization. Extensive experiments show that our approaches achieve the state-of-the-art performance on total twelve datasets from three topics, including model robustness to corruptions, domain generalization on image classification and semantic segmentation. Particularly, our $α$-BN improves 28.4\% to 43.9\% on GTA5 $\rightarrow$ Cityscapes without any training, even outperforms the latest source-free domain adaptation method.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
