Test time Adaptation through Perturbation Robustness
It addresses the challenge of dynamic real-world data for machine learning models, offering a test-time adaptation approach without changing training, which is incremental as it builds on existing domain shift methods.
The paper tackles the problem of adapting to domain shift at inference time by enforcing consistency of predictions near test samples on the image manifold, achieving performance on par or better than previous methods on benchmarks like CIFAR-10-C, CIFAR-100-C, and VisDA-C.
Data samples generated by several real world processes are dynamic in nature \textit{i.e.}, their characteristics vary with time. Thus it is not possible to train and tackle all possible distributional shifts between training and inference, using the host of transfer learning methods in literature. In this paper, we tackle this problem of adapting to domain shift at inference time \textit{i.e.}, we do not change the training process, but quickly adapt the model at test-time to handle any domain shift. For this, we propose to enforce consistency of predictions of data sampled in the vicinity of test sample on the image manifold. On a host of test scenarios like dealing with corruptions (CIFAR-10-C and CIFAR-100-C), and domain adaptation (VisDA-C), our method is at par or significantly outperforms previous methods.