α
Research
Alpha Leak
Conferences
Topics
Top Authors
Rankings
Browse All
EN
中
Home
/
Authors
/
Kwangjun Ahn
Kwangjun Ahn
11
papers
511
total citations
papers (11)
Transformers learn to implement preconditioned gradient descent for in-context learning
NEURIPS 2023
arXiv
252
citations
SGD with shuffling: optimal rates without component convexity and large epoch requirements
NEURIPS 2020
arXiv
70
citations
Efficient constrained sampling via the mirror-Langevin algorithm
NEURIPS 2021
arXiv
62
citations
The Crucial Role of Normalization in Sharpness-Aware Minimization
NEURIPS 2023
arXiv
30
citations
Reproducibility in Optimization: Theoretical Framework and Limits
NEURIPS 2022
arXiv
28
citations
Mirror Descent Maximizes Generalized Margin and Can Be Implemented Efficiently
NEURIPS 2022
arXiv
25
citations
Understanding Adam Optimizer via Online Learning of Updates: Adam is FTRL in Disguise
ICML 2024
arXiv
22
citations
How to Escape Sharp Minima with Random Perturbations
ICML 2024
arXiv
14
citations
General framework for online-to-nonconvex conversion: Schedule-free SGD is also effective for nonconvex optimization
ICML 2025
arXiv
6
citations
Through the River: Understanding the Benefit of Schedule-Free Methods for Language Model Training
NEURIPS 2025
arXiv
2
citations
Learning threshold neurons via edge of stability
NEURIPS 2023
0
citations