论文标题
TTAPS：通过使用自我安排对齐原型测试时间适应
TTAPS: Test-Time Adaption by Aligning Prototypes using Self-Supervision
论文作者
论文摘要
如今，深度神经网络在许多任务中的表现都超过了人类。但是，如果输入分布从训练中使用的分布偏离，则其性能会大大下降。最近发表的研究表明，将模型参数适应测试样本可以减轻这种性能降解。因此，在本文中，我们提出了对自我监督训练算法SHAV的新颖修改，以增加适应单个测试样品的能力。使用SWAV的提供的原型和我们派生的测试时间损失，我们将看不见的测试样品的表示与自我监督的学识渊博的原型相结合。我们在通用基准数据集CIFAR10-C上显示了我们方法的成功。
Nowadays, deep neural networks outperform humans in many tasks. However, if the input distribution drifts away from the one used in training, their performance drops significantly. Recently published research has shown that adapting the model parameters to the test sample can mitigate this performance degradation. In this paper, we therefore propose a novel modification of the self-supervised training algorithm SwAV that adds the ability to adapt to single test samples. Using the provided prototypes of SwAV and our derived test-time loss, we align the representation of unseen test samples with the self-supervised learned prototypes. We show the success of our method on the common benchmark dataset CIFAR10-C.
