Kwangjun Ahn - RankMe

62 papers A* 19B 1Journal 36Unranked 6

Year	Rank	Type	Title / Venue / Authors
2025	J	`jnl`	Dion2: A Simple Method to Shrink Matrix in Muon. CoRR Kwangjun Ahn, Noah Amsel, John Langford
2025	J	`jnl`	Dion: A Communication-Efficient Optimizer for Large Models. CoRR Kwangjun Ahn, Byron Xu
2025	A*	`conf`	Does SGD really happen in tiny subspaces? ICLR Minhak Song, Kwangjun Ahn, Chulhee Yun
2025	J	`jnl`	Efficient Joint Prediction of Multiple Future Tokens. CoRR Kwangjun Ahn, Alex Lamb, John Langford
2025	A*	`conf`	General framework for online-to-nonconvex conversion: Schedule-free SGD is also effective for nonconvex optimization. ICML Kwangjun Ahn, Gagik Magakyan, Ashok Cutkosky
2025	J	`jnl`	Next-Latent Prediction Transformers Learn Compact World Models. CoRR Jayden Teoh, Manan Tomar, Kwangjun Ahn, Edward S. Hu, Pratyusha Sharma, Riashat Islam, Alex Lamb, John Langford
2025	A*	`conf`	The Belief State Transformer. ICLR Edward S. Hu, Kwangjun Ahn, Qinghua Liu, Haoran Xu, Manan Tomar, Ada Langford, Dinesh Jayaraman, Alex Lamb, John Langford
2025	J	`jnl`	Through the River: Understanding the Benefit of Schedule-Free Methods for Language Model Training. CoRR Minhak Song, Beomhan Baek, Kwangjun Ahn, Chulhee Yun
2024	A*	`conf`	Adam with model exponential moving average is effective for nonconvex optimization. NeurIPS Kwangjun Ahn, Ashok Cutkosky
2024	J	`jnl`	Adam with model exponential moving average is effective for nonconvex optimization. CoRR Kwangjun Ahn, Ashok Cutkosky
2024	J	`jnl`	Does SGD really happen in tiny subspaces? CoRR Minhak Song, Kwangjun Ahn, Chulhee Yun
2024	J	`jnl`	General framework for online-to-nonconvex conversion: Schedule-free SGD is also effective for nonconvex optimization. CoRR Kwangjun Ahn, Gagik Magakyan, Ashok Cutkosky
2024	A*	`conf`	How to Escape Sharp Minima with Random Perturbations. ICML Kwangjun Ahn, Ali Jadbabaie, Suvrit Sra
2024	J	`jnl`	Improved Sample Complexity of Imitation Learning for Barrier Model Predictive Control. CoRR Daniel Pfrommer, Swati Padmanabhan, Kwangjun Ahn, Jack Umenberger, Tobia Marcucci, Zakaria Mhammedi, Ali Jadbabaie
2024	J	`jnl`	Learning to Achieve Goals with Belief State Transformers. CoRR Edward S. Hu, Kwangjun Ahn, Qinghua Liu, Haoran Xu, Manan Tomar, Ada Langford, Dinesh Jayaraman, Alex Lamb, John Langford
2024	A*	`conf`	Linear attention is (maybe) all you need (to understand Transformer optimization). ICLR Kwangjun Ahn, Xiang Cheng, Minhak Song, Chulhee Yun, Ali Jadbabaie, Suvrit Sra
2024	—	`conf`	On the Sample Complexity of Imitation Learning for Smoothed Model Predictive Control. CDC Daniel Pfrommer, Swati Padmanabhan, Kwangjun Ahn, Jack Umenberger, Tobia Marcucci, Zakaria Mhammedi, Ali Jadbabaie
2024	A*	`conf`	Understanding Adam Optimizer via Online Learning of Updates: Adam is FTRL in Disguise. ICML Kwangjun Ahn, Zhiyu Zhang, Yunbum Kook, Yan Dai
2024	J	`jnl`	Understanding Adam Optimizer via Online Learning of Updates: Adam is FTRL in Disguise. CoRR Kwangjun Ahn, Zhiyu Zhang, Yunbum Kook, Yan Dai
2023	J	`jnl`	A Unified Approach to Controlling Implicit Regularization via Mirror Descent. CoRR Haoyuan Sun, Khashayar Gatmiry, Kwangjun Ahn, Navid Azizan
2023	J	`jnl`	A Unified Approach to Controlling Implicit Regularization via Mirror Descent. J. Mach. Learn. Res. Haoyuan Sun, Khashayar Gatmiry, Kwangjun Ahn, Navid Azizan
2023	J	`jnl`	How to escape sharp minima. CoRR Kwangjun Ahn, Ali Jadbabaie, Suvrit Sra
2023	A*	`conf`	Learning threshold neurons via edge of stability. NeurIPS Kwangjun Ahn, Sébastien Bubeck, Sinho Chewi, Yin Tat Lee, Felipe Suarez, Yi Zhang
2023	J	`jnl`	Linear attention is (maybe) all you need (to understand transformer optimization). CoRR Kwangjun Ahn, Xiang Cheng, Minhak Song, Chulhee Yun, Ali Jadbabaie, Suvrit Sra
2023	—	`conf`	Model Predictive Control via On-Policy Imitation Learning. L4DC Kwangjun Ahn, Zakaria Mhammedi, Horia Mania, Zhang-Wei Hong, Ali Jadbabaie
2023	J	`jnl`	Smooth Model Predictive Control with Applications to Statistical Learning. CoRR Kwangjun Ahn, Daniel Pfrommer, Jack Umenberger, Tobia Marcucci, Zakaria Mhammedi, Ali Jadbabaie
2023	A*	`conf`	The Crucial Role of Normalization in Sharpness-Aware Minimization. NeurIPS Yan Dai, Kwangjun Ahn, Suvrit Sra
2023	J	`jnl`	The Crucial Role of Normalization in Sharpness-Aware Minimization. CoRR Yan Dai, Kwangjun Ahn, Suvrit Sra
2023	A*	`conf`	Transformers learn to implement preconditioned gradient descent for in-context learning. NeurIPS Kwangjun Ahn, Xiang Cheng, Hadi Daneshmand, Suvrit Sra
2023	J	`jnl`	Transformers learn to implement preconditioned gradient descent for in-context learning. CoRR Kwangjun Ahn, Xiang Cheng, Hadi Daneshmand, Suvrit Sra
2022	A*	`conf`	Agnostic Learnability of Halfspaces via Logistic Loss. ICML Ziwei Ji, Kwangjun Ahn, Pranjal Awasthi, Satyen Kale, Stefani Karp
2022	J	`jnl`	Agnostic Learnability of Halfspaces via Logistic Loss. CoRR Ziwei Ji, Kwangjun Ahn, Pranjal Awasthi, Satyen Kale, Stefani Karp
2022	J	`jnl`	Learning threshold neurons via the "edge of stability". CoRR Kwangjun Ahn, Sébastien Bubeck, Sinho Chewi, Yin Tat Lee, Felipe Suarez, Yi Zhang
2022	A*	`conf`	Mirror Descent Maximizes Generalized Margin and Can Be Implemented Efficiently. NeurIPS Haoyuan Sun, Kwangjun Ahn, Christos Thrampoulidis, Navid Azizan
2022	J	`jnl`	Mirror Descent Maximizes Generalized Margin and Can Be Implemented Efficiently. CoRR Haoyuan Sun, Kwangjun Ahn, Christos Thrampoulidis, Navid Azizan
2022	J	`jnl`	Model Predictive Control via On-Policy Imitation Learning. CoRR Kwangjun Ahn, Zakaria Mhammedi, Horia Mania, Zhang-Wei Hong, Ali Jadbabaie
2022	—	`conf`	One-Pass Learning via Bridging Orthogonal Gradient Descent and Recursive Least-Squares. CDC Youngjae Min, Kwangjun Ahn, Navid Azizan
2022	J	`jnl`	One-Pass Learning via Bridging Orthogonal Gradient Descent and Recursive Least-Squares. CoRR Youngjae Min, Kwangjun Ahn, Navid Azizan
2022	A*	`conf`	Reproducibility in Optimization: Theoretical Framework and Limits. NeurIPS Kwangjun Ahn, Prateek Jain, Ziwei Ji, Satyen Kale, Praneeth Netrapalli, Gil I. Shamir
2022	J	`jnl`	Reproducibility in Optimization: Theoretical Framework and Limits. CoRR Kwangjun Ahn, Prateek Jain, Ziwei Ji, Satyen Kale, Praneeth Netrapalli, Gil I. Shamir
2022	—	`conf`	Understanding Nesterov's Acceleration via Proximal Point Method. SOSA Kwangjun Ahn, Suvrit Sra
2022	A*	`conf`	Understanding the unstable convergence of gradient descent. ICML Kwangjun Ahn, Jingzhao Zhang, Suvrit Sra
2022	J	`jnl`	Understanding the unstable convergence of gradient descent. CoRR Kwangjun Ahn, Jingzhao Zhang, Suvrit Sra
2021	A*	`conf`	Efficient constrained sampling via the mirror-Langevin algorithm. NeurIPS Kwangjun Ahn, Sinho Chewi
2021	A*	`conf`	Optimal dimension dependence of the Metropolis-Adjusted Langevin Algorithm. COLT Sinho Chewi, Chen Lu, Kwangjun Ahn, Xiang Cheng, Thibaut Le Gouic, Philippe Rigollet
2021	J	`jnl`	Riemannian Perspective on Matrix Factorization. CoRR Kwangjun Ahn, Felipe Suarez
2020	—	`conf`	A Simpler Strong Refutation of Random k-XOR. APPROX-RANDOM Kwangjun Ahn
2020	J	`jnl`	A simpler strong refutation of random k-XOR. CoRR Kwangjun Ahn
2020	J	`jnl`	Efficient constrained sampling via the mirror-Langevin algorithm. CoRR Kwangjun Ahn, Sinho Chewi
2020	A*	`conf`	From Nesterov's Estimate Sequence to Riemannian Acceleration. COLT Kwangjun Ahn, Suvrit Sra
2020	J	`jnl`	From Proximal Point Method to Nesterov's Acceleration. CoRR Kwangjun Ahn
2020	J	`jnl`	On Tight Convergence Rates of Without-replacement SGD. CoRR Kwangjun Ahn, Suvrit Sra
2020	J	`jnl`	Optimal dimension dependence of the Metropolis-Adjusted Langevin Algorithm. CoRR Sinho Chewi, Chen Lu, Kwangjun Ahn, Xiang Cheng, Thibaut Le Gouic, Philippe Rigollet
2020	A*	`conf`	SGD with shuffling: optimal rates without component convexity and large epoch requirements. NeurIPS Kwangjun Ahn, Chulhee Yun, Suvrit Sra
2019	J	`jnl`	Community Recovery in Hypergraphs. IEEE Trans. Inf. Theory Kwangjun Ahn, Kangwook Lee, Changho Suh
2018	A*	`conf`	Binary Rating Estimation with Graph Side Information. NeurIPS Kwangjun Ahn, Kangwook Lee, Hyunseung Cha, Changho Suh
2018	J	`jnl`	Hypergraph Spectral Clustering in the Weighted Stochastic Block Model. CoRR Kwangjun Ahn, Kangwook Lee, Changho Suh
2018	J	`jnl`	Hypergraph Spectral Clustering in the Weighted Stochastic Block Model. IEEE J. Sel. Top. Signal Process. Kwangjun Ahn, Kangwook Lee, Changho Suh
2017	J	`jnl`	Community Recovery in Hypergraphs. CoRR Kwangjun Ahn, Kangwook Lee, Changho Suh
2017	J	`jnl`	Computing the maximum matching width is NP-hard. CoRR Kwangjun Ahn, Jisu Jeong
2017	B	`conf`	Information-theoretic limits of subspace clustering. ISIT Kwangjun Ahn, Kangwook Lee, Changho Suh
2016	—	`conf`	Community recovery in hypergraphs. Allerton Kwangjun Ahn, Kangwook Lee, Changho Suh