Exploring Target Representations for Masked Autoencoders
This work addresses the challenge of simplifying self-supervised pre-training for computer vision researchers, though it is incremental in refining existing masked autoencoder paradigms.
The paper tackles the problem of target representation selection in masked autoencoders for self-supervised visual learning, showing that careful target design is unnecessary and proposing a multi-stage masked distillation method that outperforms previous methods by nontrivial margins on tasks like classification and object detection.
Masked autoencoders have become popular training paradigms for self-supervised visual representation learning. These models randomly mask a portion of the input and reconstruct the masked portion according to the target representations. In this paper, we first show that a careful choice of the target representation is unnecessary for learning good representations, since different targets tend to derive similarly behaved models. Driven by this observation, we propose a multi-stage masked distillation pipeline and use a randomly initialized model as the teacher, enabling us to effectively train high-capacity models without any efforts to carefully design target representations. Interestingly, we further explore using teachers of larger capacity, obtaining distilled students with remarkable transferring ability. On different tasks of classification, transfer learning, object detection, and semantic segmentation, the proposed method to perform masked knowledge distillation with bootstrapped teachers (dBOT) outperforms previous self-supervised methods by nontrivial margins. We hope our findings, as well as the proposed method, could motivate people to rethink the roles of target representations in pre-training masked autoencoders.The code and pre-trained models are publicly available at https://github.com/liuxingbin/dbot.