Zeyuan Allen-Zhu

70 papers A* 33Journal 36Unranked 1

Year	Rank	Type	Title / Venue / Authors
2025	J	`jnl`	Making Hard Problems Easier with Custom Data Distributions and Loss Regularization: A Case Study in Modular Arithmetic. IACR Cryptol. ePrint Arch. Eshika Saxena, Alberto Alfarano, François Charton, Zeyuan Allen-Zhu, Emily Wenger, Kristin E. Lauter
2025	J	`jnl`	Physics of Language Models: Part 1, Learning Hierarchical Language Structures. Trans. Mach. Learn. Res. Zeyuan Allen-Zhu, Yuanzhi Li
2025	A*	`conf`	Physics of Language Models: Part 2.1, Grade-School Math and the Hidden Reasoning Process. ICLR Tian Ye, Zicheng Xu, Yuanzhi Li, Zeyuan Allen-Zhu
2025	A*	`conf`	Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problems. ICLR Tian Ye, Zicheng Xu, Yuanzhi Li, Zeyuan Allen-Zhu
2025	A*	`conf`	Physics of Language Models: Part 3.2, Knowledge Manipulation. ICLR Zeyuan Allen-Zhu, Yuanzhi Li
2025	A*	`conf`	Physics of Language Models: Part 3.3, Knowledge Capacity Scaling Laws. ICLR Zeyuan Allen-Zhu, Yuanzhi Li
2025	J	`jnl`	Physics of Language Models: Part 4.1, Architecture Design and the Magic of Canon Layers. CoRR Zeyuan Allen-Zhu
2024	J	`jnl`	Physics of Language Models: Part 2.1, Grade-School Math and the Hidden Reasoning Process. CoRR Tian Ye, Zicheng Xu, Yuanzhi Li, Zeyuan Allen-Zhu
2024	J	`jnl`	Physics of Language Models: Part 2.2, How to Learn From Mistakes on Grade-School Math Problems. CoRR Tian Ye, Zicheng Xu, Yuanzhi Li, Zeyuan Allen-Zhu
2024	A*	`conf`	Physics of Language Models: Part 3.1, Knowledge Storage and Extraction. ICML Zeyuan Allen-Zhu, Yuanzhi Li
2024	J	`jnl`	Physics of Language Models: Part 3.3, Knowledge Capacity Scaling Laws. CoRR Zeyuan Allen-Zhu, Yuanzhi Li
2024	J	`jnl`	Reverse Training to Nurse the Reversal Curse. CoRR Olga Golovneva, Zeyuan Allen-Zhu, Jason Weston, Sainbayar Sukhbaatar
2023	A*	`conf`	Backward Feature Correction: How Deep Learning Performs Deep (Hierarchical) Learning. COLT Zeyuan Allen-Zhu, Yuanzhi Li
2023	A*	`conf`	Forward Super-Resolution: How Can GANs Learn Hierarchical Generative Models for Real-World Distributions. ICLR Zeyuan Allen-Zhu, Yuanzhi Li
2023	J	`jnl`	Physics of Language Models: Part 1, Context-Free Grammar. CoRR Zeyuan Allen-Zhu, Yuanzhi Li
2023	J	`jnl`	Physics of Language Models: Part 3.2, Knowledge Manipulation. CoRR Zeyuan Allen-Zhu, Yuanzhi Li
2023	A*	`conf`	SALSA VERDE: a machine learning attack on LWE with sparse small secrets. NeurIPS Cathy Yuanchen Li, Emily Wenger, Zeyuan Allen-Zhu, François Charton, Kristin E. Lauter
2023	J	`jnl`	SALSA VERDE: a machine learning attack on Learning With Errors with sparse small secrets. CoRR Cathy Yuanchen Li, Jana Sotáková, Emily Wenger, Zeyuan Allen-Zhu, François Charton, Kristin E. Lauter
2023	J	`jnl`	SALSA VERDE: a machine learning attack on Learning with Errors with sparse small secrets. IACR Cryptol. ePrint Arch. Cathy Yuanchen Li, Jana Sotáková, Emily Wenger, Zeyuan Allen-Zhu, François Charton, Kristin E. Lauter
2023	A*	`conf`	Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learning. ICLR Zeyuan Allen-Zhu, Yuanzhi Li
2022	A*	`conf`	LoRA: Low-Rank Adaptation of Large Language Models. ICLR Edward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, Weizhu Chen
2021	A*	`conf`	Byzantine-Resilient Non-Convex Stochastic Gradient Descent. ICLR Zeyuan Allen-Zhu, Faeze Ebrahimianghazani, Jerry Li, Dan Alistarh
2021	A*	`conf`	Feature Purification: How Adversarial Training Performs Robust Deep Learning. FOCS Zeyuan Allen-Zhu, Yuanzhi Li
2021	J	`jnl`	Forward Super-Resolution: How Can GANs Learn Hierarchical Generative Models for Real-World Distributions. CoRR Zeyuan Allen-Zhu, Yuanzhi Li
2021	J	`jnl`	LoRA: Low-Rank Adaptation of Large Language Models. CoRR Edward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Weizhu Chen
2021	J	`jnl`	Near-optimal discrete optimization for experimental design: a regret minimization approach. Math. Program. Zeyuan Allen-Zhu, Yuanzhi Li, Aarti Singh, Yining Wang
2020	J	`jnl`	Backward Feature Correction: How Deep Learning Performs Deep Learning. CoRR Zeyuan Allen-Zhu, Yuanzhi Li
2020	J	`jnl`	Byzantine-Resilient Non-Convex Stochastic Gradient Descent. CoRR Zeyuan Allen-Zhu, Faeze Ebrahimian, Jerry Li, Dan Alistarh
2020	J	`jnl`	Feature Purification: How Adversarial Training Performs Robust Deep Learning. CoRR Zeyuan Allen-Zhu, Yuanzhi Li
2020	J	`jnl`	Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learning. CoRR Zeyuan Allen-Zhu, Yuanzhi Li
2019	A*	`conf`	A Convergence Theory for Deep Learning via Over-Parameterization. ICML Zeyuan Allen-Zhu, Yuanzhi Li, Zhao Song
2019	A*	`conf`	Can SGD Learn Recurrent Neural Networks with Provable Generalization? NeurIPS Zeyuan Allen-Zhu, Yuanzhi Li
2019	J	`jnl`	Can SGD Learn Recurrent Neural Networks with Provable Generalization? CoRR Zeyuan Allen-Zhu, Yuanzhi Li
2019	A*	`conf`	Learning and Generalization in Overparameterized Neural Networks, Going Beyond Two Layers. NeurIPS Zeyuan Allen-Zhu, Yuanzhi Li, Yingyu Liang
2019	J	`jnl`	Nearly linear-time packing and covering LP solvers - Achieving width-independence and -convergence. Math. Program. Zeyuan Allen-Zhu, Lorenzo Orecchia
2019	A*	`conf`	On the Convergence Rate of Training Recurrent Neural Networks. NeurIPS Zeyuan Allen-Zhu, Yuanzhi Li, Zhao Song
2019	J	`jnl`	The Lingering of Gradients: How to Reuse Gradients over Time. CoRR Zeyuan Allen-Zhu, David Simchi-Levi, Xinshang Wang
2019	A*	`conf`	What Can ResNet Learn Efficiently, Going Beyond Kernels? NeurIPS Zeyuan Allen-Zhu, Yuanzhi Li
2019	J	`jnl`	What Can ResNet Learn Efficiently, Going Beyond Kernels? CoRR Zeyuan Allen-Zhu, Yuanzhi Li
2018	J	`jnl`	A Convergence Theory for Deep Learning via Over-Parameterization. CoRR Zeyuan Allen-Zhu, Yuanzhi Li, Zhao Song
2018	A*	`conf`	Byzantine Stochastic Gradient Descent. NeurIPS Dan Alistarh, Zeyuan Allen-Zhu, Jerry Li
2018	J	`jnl`	Byzantine Stochastic Gradient Descent. CoRR Dan Alistarh, Zeyuan Allen-Zhu, Jerry Li
2018	J	`jnl`	How To Make the Gradients Small Stochastically. CoRR Zeyuan Allen-Zhu
2018	A*	`conf`	How To Make the Gradients Small Stochastically: Even Faster Convex and Nonconvex SGD. NeurIPS Zeyuan Allen-Zhu
2018	A*	`conf`	Is Q-Learning Provably Efficient? NeurIPS Chi Jin, Zeyuan Allen-Zhu, Sébastien Bubeck, Michael I. Jordan
2018	J	`jnl`	Is Q-learning Provably Efficient? CoRR Chi Jin, Zeyuan Allen-Zhu, Sébastien Bubeck, Michael I. Jordan
2018	A*	`conf`	Katyusha X: Practical Momentum Method for Stochastic Sum-of-Nonconvex Optimization. ICML Zeyuan Allen-Zhu
2018	J	`jnl`	Katyusha X: Practical Momentum Method for Stochastic Sum-of-Nonconvex Optimization. CoRR Zeyuan Allen-Zhu
2018	J	`jnl`	Learning and Generalization in Overparameterized Neural Networks, Going Beyond Two Layers. CoRR Zeyuan Allen-Zhu, Yuanzhi Li, Yingyu Liang
2018	A*	`conf`	Make the Minority Great Again: First-Order Regret Bound for Contextual Bandits. ICML Zeyuan Allen-Zhu, Sébastien Bubeck, Yuanzhi Li
2018	J	`jnl`	Make the Minority Great Again: First-Order Regret Bound for Contextual Bandits. CoRR Zeyuan Allen-Zhu, Sébastien Bubeck, Yuanzhi Li
2018	A*	`conf`	NEON2: Finding Local Minima via First-Order Oracles. NeurIPS Zeyuan Allen-Zhu, Yuanzhi Li
2018	A*	`conf`	Natasha 2: Faster Non-Convex Optimization Than SGD. NeurIPS Zeyuan Allen-Zhu
2018	J	`jnl`	On the Convergence Rate of Training Recurrent Neural Networks. CoRR Zeyuan Allen-Zhu, Yuanzhi Li, Zhao Song
2018	J	`jnl`	Operator Scaling via Geodesically Convex Optimization, Invariant Theory and Polynomial Identity Testing. CoRR Zeyuan Allen-Zhu, Ankit Garg, Yuanzhi Li, Rafael Mendes de Oliveira, Avi Wigderson
2018	A*	`conf`	Operator scaling via geodesically convex optimization, invariant theory and polynomial identity testing. STOC Zeyuan Allen-Zhu, Ankit Garg, Yuanzhi Li, Rafael Mendes de Oliveira, Avi Wigderson
2018	A*	`conf`	The Lingering of Gradients: How to Reuse Gradients Over Time. NeurIPS Zeyuan Allen-Zhu, David Simchi-Levi, Xinshang Wang
2017	A*	`conf`	Doubly Accelerated Methods for Faster CCA and Generalized Eigendecomposition. ICML Zeyuan Allen-Zhu, Yuanzhi Li
2017	A*	`conf`	Faster Principal Component Regression and Stable Matrix Chebyshev Approximation. ICML Zeyuan Allen-Zhu, Yuanzhi Li
2017	A*	`conf`	First Efficient Convergence for Streaming k-PCA: A Global, Gap-Free, and Near-Optimal Rate. FOCS Zeyuan Allen-Zhu, Yuanzhi Li
2017	A*	`conf`	Follow the Compressed Leader: Faster Online Learning of Eigenvectors and Faster MMWU. ICML Zeyuan Allen-Zhu, Yuanzhi Li
2017	J	`jnl`	Katyusha: The First Direct Acceleration of Stochastic Gradient Methods. J. Mach. Learn. Res. Zeyuan Allen-Zhu
2017	—	`conf`	Linear Convergence of a Frank-Wolfe Type Algorithm over Trace-Norm Balls. NIPS Zeyuan Allen-Zhu, Elad Hazan, Wei Hu, Yuanzhi Li
2017	J	`jnl`	Linear Convergence of a Frank-Wolfe Type Algorithm over Trace-Norm Balls. CoRR Zeyuan Allen-Zhu, Elad Hazan, Wei Hu, Yuanzhi Li
2017	A*	`conf`	Much Faster Algorithms for Matrix Scaling. FOCS Zeyuan Allen-Zhu, Yuanzhi Li, Rafael Mendes de Oliveira, Avi Wigderson
2017	J	`jnl`	Natasha 2: Faster Non-Convex Optimization Than SGD. CoRR Zeyuan Allen-Zhu
2017	A*	`conf`	Natasha: Faster Non-Convex Stochastic Optimization via Strongly Non-Convex Parameter. ICML Zeyuan Allen-Zhu
2017	A*	`conf`	Near-Optimal Design of Experiments via Regret Minimization. ICML Zeyuan Allen-Zhu, Yuanzhi Li, Aarti Singh, Yining Wang
2017	J	`jnl`	Near-Optimal Discrete Optimization for Experimental Design: A Regret Minimization Approach. CoRR Zeyuan Allen-Zhu, Yuanzhi Li, Aarti Singh, Yining Wang
2017	J	`jnl`	Neon2: Finding Local Minima via First-Order Oracles. CoRR Zeyuan Allen-Zhu, Yuanzhi Li