In context of Test-time Adaptation(TTA), we propose a regularizer, dubbed
Gradient Alignment with Prototype feature (GAP), which alleviates the
inappropriate guidance from entropy minimization loss from misclassified pseudo
label. We developed a gradient alignment loss to precisely manage the
adaptation process, ensuring that changes made for some data don't negatively
impact the model's performance on other data. We introduce a prototype feature
of a class as a proxy measure of the negative impact. To make GAP regularizer
feasible under the TTA constraints, where model can only access test data
without labels, we tailored its formula in two ways: approximating prototype
features with weight vectors of the classifier, calculating gradient without
back-propagation. We demonstrate GAP significantly improves TTA methods across
various datasets, which proves its versatility and effectiveness.

通过提出一种称为原型特征的梯度对齐正则化器（GAP）来减轻测试时自适应中由于错误分类的伪标签的信息熵最小化损失的不适当指导。我们开发了一种梯度对齐损失来精确地管理自适应过程，确保对一些数据所做的更改不会对其他数据上的模型性能产生负面影响。为了使 GAP 正则化器能够在只能访问没有标签的测试数据的 TTA 约束下实现，我们通过两种方式调整了其公式：用分类器的权重向量近似原型特征，计算梯度而无需反向传播。我们展示了 GAP 显著改善了各种数据集上的 TTA 方法，证明了其多功能性和有效性。