Paper ID: 2209.07522
Test-Time Training with Masked Autoencoders
Yossi Gandelsman, Yu Sun, Xinlei Chen, Alexei A. Efros
Test-time training adapts to a new test distribution on the fly by optimizing a model for each test input using self-supervision. In this paper, we use masked autoencoders for this one-sample learning problem. Empirically, our simple method improves generalization on many visual benchmarks for distribution shifts. Theoretically, we characterize this improvement in terms of the bias-variance trade-off.
Submitted: Sep 15, 2022