Understanding Masked Autoencoders From a Local Contrastive Perspective
This work provides incremental insights into MAE's efficacy for researchers in self-supervised learning, potentially inspiring new methods.
The paper tackled the problem of understanding the mechanisms behind Masked AutoEncoder (MAE) in self-supervised learning by proposing a local contrastive perspective and introducing LC-MAE to analyze its reconstructive and contrastive aspects, revealing insights into invariance learning and the roles of the decoder and random masking.
Masked AutoEncoder (MAE) has revolutionized the field of self-supervised learning with its simple yet effective masking and reconstruction strategies. However, despite achieving state-of-the-art performance across various downstream vision tasks, the underlying mechanisms that drive MAE's efficacy are less well-explored compared to the canonical contrastive learning paradigm. In this paper, we first propose a local perspective to explicitly extract a local contrastive form from MAE's reconstructive objective at the patch level. And then we introduce a new empirical framework, called Local Contrastive MAE (LC-MAE), to analyze both reconstructive and contrastive aspects of MAE. LC-MAE reveals that MAE learns invariance to random masking and ensures distribution consistency between the learned token embeddings and the original images. Furthermore, we dissect the contribution of the decoder and random masking to MAE's success, revealing both the decoder's learning mechanism and the dual role of random masking as data augmentation and effective receptive field restriction. Our experimental analysis sheds light on the intricacies of MAE and summarizes some useful design methodologies, which can inspire more powerful visual self-supervised methods.