Louis-Philippe Morency, Kun Zhang, Eric Xing, Lingjing Kong, Guangyi Chen, Yuejie Chi, Martin Ma
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Masked autoencoder (MAE), a simple and effective selfsupervised learning framework based on the reconstruction of masked image regions, has recently achieved prominent success in a variety of vision tasks. Despite the emergence of intriguing empirical observations on MAE, a theoretically principled understanding is still lacking. In this work, we formally characterize and justify existing empirical insights and provide theoretical guarantees of MAE. We formulate the underlying data-generating process as a hierarchical latent variable model, and show that under reasonable assumptions, MAE provably identifies a set of latent variables in the hierarchical model, explaining why MAE can extract high-level information from pixels. Further, we show how key hyperparameters in MAE (the masking ratio and the patch size) determine which true latent variables to be recovered, therefore influencing the level of semantic information in the representation. Specifically, extremely large or small masking ratios inevitably lead to low-level representations. Our theory offers coherent explanations of existing empirical observations and provides insights for potential empirical improvements and fundamental limitations of the masked-reconstruction paradigm. We conduct extensive experiments to validate our theoretical insights.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2306.04898")
get_code_for_paper("2306.04898")
have("2306.04898")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2306.04898.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2306.04898)
Connect an agent — have() is free.