notesum.ai

Published at May 13

Unveil Benign Overfitting for Transformer in Vision: Training Dynamics, Convergence, and Generalization

NeurIPS

Released Date: May 13, 2024

Authors: Jiarui Jiang1, Wei Huang2, Miao Zhang1, Taiji Suzuki3, Liqiang Nie1

Aff.: 1Harbin Institute of Technology, Shenzhen; 2RIKEN AIP; 3University of Tokyo

Arxiv: https://openreview.net/pdf/a150caf3828f329dc59605cea4924ef3af84946d.pdf