Test-Time-Training (TTT) is an approach to cope with out-of-distribution
(OOD) data by adapting a trained model to distribution shifts occurring at
test-time. We propose to perform this adaptation via Activation Matching
(ActMAD): We analyze activations of the model and align activation statistics
of the OOD test data to those of the training data. In contrast to existing
methods, which model the distribution of entire channels in the ultimate layer
of the feature extractor, we model the distribution of each feature in multiple
layers across the network. This results in a more fine-grained supervision and
makes ActMAD attain state of the art performance on CIFAR-100C and Imagenet-C.
ActMAD is also architecture- and task-agnostic, which lets us go beyond image
classification, and score 15.4% improvement over previous approaches when
evaluating a KITTI-trained object detector on KITTI-Fog. Our experiments
highlight that ActMAD can be applied to online adaptation in realistic
scenarios, requiring little data to attain its full performance.

本文提出了基于 Activation Matching（ActMAD） 实现的 Test-Time-Training（TTT）方法，通过分析神经网络的激活函数并调节其分布，来实现对测试数据及时的辨识和适应。与现有的方法不同的是，ActMAD 不仅对神经网络最末端而且对多层特征的分布进行建模，更细粒度的监督增强了其性能，在 CIFAR-100C 和 Imagenet-C 数据集上实现了状态 - 最佳的结果，并且在 KITTI-Fog 上获得了比以前的方法高 15.4% 的成果，这表明 ActMAD 可以在线应用于实际场景的自适应问题