Kwangjun Ahn
Massachusetts Institute of Technology
H-index: 12
North America-United States
Top articles of Kwangjun Ahn
Understanding Adam Optimizer via Online Learning of Updates: Adam is FTRL in Disguise
arXiv preprint arXiv:2402.01567
2024/2/2
Smooth Model Predictive Control with Applications to Statistical Learning
arXiv preprint arXiv:2306.01914
2023/6/2
Transformers learn to implement preconditioned gradient descent for in-context learning
Advances in Neural Information Processing Systems
2024/2/13
The Crucial Role of Normalization in Sharpness-Aware Minimization
Advances in Neural Information Processing Systems
2024/2/13
SpecTr++: Improved transport plans for speculative decoding of large language models
2023/12/28
Kwangjun Ahn
H-Index: 6
Ziteng Sun
H-Index: 9
Linear attention is (maybe) all you need (to understand transformer optimization)
2023/10/2
Kwangjun Ahn
H-Index: 6
Xiang Cheng
H-Index: 7
Chulhee Yun
H-Index: 7
Ali Jadbabaie
H-Index: 47
Suvrit Sra
H-Index: 42
A Unified Approach to Controlling Implicit Regularization via Mirror Descent
Journal of Machine Learning Research (JMLR)
2023/6/24
Haoyuan Sun
H-Index: 1
Kwangjun Ahn
H-Index: 6
Model Predictive Control via On-Policy Imitation Learning
arXiv preprint arXiv:2210.09206
2022/10/17
Learning threshold neurons via the "edge of stability"
arxiv.org/2212.07469
2022/12/14
One-Pass Learning via Bridging Orthogonal Gradient Descent and Recursive Least-Squares
2022/7/28
Youngjae Min
H-Index: 3
Kwangjun Ahn
H-Index: 6
Mirror Descent Maximizes Generalized Margin and Can Be Implemented Efficiently
2022/5/25
Haoyuan Sun
H-Index: 1
Kwangjun Ahn
H-Index: 6
Understanding the unstable convergence of gradient descent
ICML 2022 (arXiv:2204.01050)
2022/4/3
Reproducibility in Optimization: Theoretical Framework and Limits
NeurIPS 2022
2022/2/9
Agnostic Learnability of Halfspaces via Logistic Loss
2022/6/28
Understanding Nesterov's Acceleration via Proximal Point Method
2022
Kwangjun Ahn
H-Index: 6
Suvrit Sra
H-Index: 42
Optimal dimension dependence of the metropolis-adjusted langevin algorithm
2021/7/21
Sinho Chewi
H-Index: 3
Chen Lu
H-Index: 6
Kwangjun Ahn
H-Index: 6
Xiang Cheng
H-Index: 7
Philippe Rigollet
H-Index: 27
Efficient constrained sampling via the mirror-Langevin algorithm
2021/5/21
Kwangjun Ahn
H-Index: 6
Sinho Chewi
H-Index: 3
Riemannian perspective on matrix factorization
arXiv preprint arXiv:2102.00937
2021/2/1
Kwangjun Ahn
H-Index: 6
From Proximal Point Method to Accelerated Methods on Riemannian Manifolds
2021
Kwangjun Ahn
H-Index: 6
A simpler strong refutation of random -XOR
International Conference on Randomization and Computation (APPROX/RANDOM) 2020,
2020/8/8
Kwangjun Ahn
H-Index: 6