all AI news
Residual Connections Harm Self-Supervised Abstract Feature Learning
April 18, 2024, 4:43 a.m. | Xiao Zhang, Ruoxi Jiang, William Gao, Rebecca Willett, Michael Maire
cs.CV updates on arXiv.org arxiv.org
Abstract: We demonstrate that adding a weighting factor to decay the strength of identity shortcuts within residual networks substantially improves semantic feature learning in the state-of-the-art self-supervised masked autoencoding (MAE) paradigm. Our modification to the identity shortcuts within a VIT-B/16 backbone of an MAE boosts linear probing accuracy on ImageNet from 67.3% to 72.3%. This significant gap suggests that, while residual connection structure serves an essential role in facilitating gradient propagation, it may have a harmful …
abstract accuracy art arxiv cs.cv feature harm identity linear networks paradigm residual semantic state type vit
More from arxiv.org / cs.CV updates on arXiv.org
Jobs in AI, ML, Big Data
AI Research Scientist
@ Vara | Berlin, Germany and Remote
Data Architect
@ University of Texas at Austin | Austin, TX
Data ETL Engineer
@ University of Texas at Austin | Austin, TX
Lead GNSS Data Scientist
@ Lurra Systems | Melbourne
Senior Machine Learning Engineer (MLOps)
@ Promaton | Remote, Europe
Senior Machine Learning Engineer
@ Samsara | Canada - Remote