2025

• Subspace Optimization for Large Language Models with Convergence Guarantees
  Y. He, P. Li, Y. Hu, C. Chen, and K. Yuan
  International Conference on Machine Learning (ICML)
  

• A Memory Efficient Randomized Subspace Optimization Method for Training Large Language Models
  Y. Chen, Y. Zhang, Y. Liu, K. Yuan, and Z. Wen
  International Conference on Machine Learning (ICML)
  

• Distributed Retraction-Free and Communication-Efficient Optimization on the Stiefel Manifold
  Y. Song, P. Li, B. Gao, and K. Yuan
  International Conference on Machine Learning (ICML)
  

• Achieving Linear Speedup and Optimal Complexity for Decentralized Optimization over Row-stochastic Networks
  L. Liang, G. Luo, X. Chen, and K. Yuan
  International Conference on Machine Learning (ICML)   Spotlight

• Efficient Multi-Objective Learning under Preference Guidance: A First-Order Penalty Approach
  L. Chen, Q. Xiao, E. H. Fukuda, X. Chen, K. Yuan, and T. Chen
  International Conference on Machine Learning (ICML)   Spotlight

• Understanding the Influence of Digraphs on Decentralized Optimization: Effective Metrics, Lower Bound, and Optimal Algorithm
  L. Liang, X. Huang, R. Xin, K. Yuan
  SIAM Journal on Optimization
  

• BEVHeight++: Toward Robust Visual Centric 3D Object Detection
  L. Yang, T. Tang, J. Li, K. Yuan, K. Wu, P. Chen, L. Wang, Y. Huang, L. Li, X. Zhang, K. Yu
  IEEE Transactions on Pattern Analysis and Machine Intelligence
  

• CE-LoRA: Computation-Efficient LoRA Fine-Tuning for Language Models
  G. Chen, Y. He, Y. Hu, K. Yuan, and B. Yuan
  arXiv preprint: 2502.01378
  

• Enhancing Zeroth-Order Fine-Tuning for Language Models with Low-Rank Structures
  Y. Chen, Y. Zhang, L. Cao, K. Yuan, and Z. Wen
  International Conference on Learning Representations (ICLR)
  

2024

• Heavy-Tail phenomenon in decentralized SGD
  M. Gurbuzbalaban, Y. Hu, U. Simsekli, K. Yuan, and L. Zhu
  IISE Transactions
  

• A Mathematics-Inspired Learning-to-Optimize Framework for Decentralized Optimization
  Y. He, Q. Shang, X. Huang, J. Liu, and K. Yuan
  arXiv preprint: 2410.01700
  

• SPARKLE: A Unified Single-Loop Primal-Dual Framework for Decentralized Bilevel Optimization
  S. Zhu, B. Kong, S. Lu, X. Huang, and K. Yuan
  The Conference on Neural Information Processing Systems (NeurIPS)
  

• S3 Attention: Improving Long Sequence Attention with Smoothed Skeleton Sketching
  X. Wang, T. Zhou, J. Zhu, J. Liu, K. Yuan, T. Yao, W. Yin, R. Jin, H. Cai
  IEEE Journal of Selected Topics in Signal Processing
  

• Distributed Bilevel Optimization with Communication Compression
  Y. He, J. Hu, X. Huang, S. Lu, B. Wang, and K. Yuan
  International Conference on Machine Learning (ICML)
  

• Asynchronous Diffusion Learning with Agent Subsampling and Local Updates
  Elsa Rizk, Kun Yuan, Ali H. Sayed
  The IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

• Momentum Benefits Non-IID Federated Learning Simply and Provably
  Z. Cheng , X. Huang, P. Wu, and K. Yuan
  International Conference on Learning Representations (ICLR)
  

• Decentralized Bilevel Optimization over Graphs: Loopless Algorithmic Update and Transient Iteration Complexity
  B. Kong, S. Zhu, S. Lu, X. Huang, K. Yuan
  arXiv preprint: 2402.03167
  

2023

• Sharper Convergence Guarantees for Federated Learning with Partial Model Personalization
  Y. Chen, L. Cao, K. Yuan, and Z. Wen
  arXiv preprint: 2309.17409
  

• Lower Bounds and Accelerated Algorithms in Distributed Stochastic Optimization with Communication Compression
  Y. He , X. Huang, Y. Chen, W. Yin, and K. Yuan
  arXiv preprint: 2305.07612
  

• An Enhanced Gradient-Tracking Bound for Distributed Online Stochastic Convex Optimization
  S. A. Alghunaim and K. Yuan
  Signal Processing

• Unbiased Compression Saves Communication in Distributed Optimization: When and How Much?
  Y. He , X. Huang, and K. Yuan
  The Conference on Neural Information Processing Systems (NeurIPS)
  

• Removing data heterogeneity influence enhances network topology dependence of decentralized SGD
  K. Yuan, S. A. Alghunaim, and X. Huang
  Journal of Machine Learning Research (JMLR)

• Achieving Linear Speedup with Network-Independent Learning Rates in Decentralized Stochastic Optimization
  H. Yuan, S. A. Alghunaim, and K. Yuan
  IEEE Conference on Decision and Control (CDC)

• On the Performance of Gradient Tracking with Local Updates
  E. D. H. Nguyen, S. A. Alghunaim, K. Yuan, and C. A. Uribe
  IEEE Conference on Decision and Control (CDC)

• DSGD-CECA: Decentralized SGD with Communication-Optimal Exact Consensus Algorithm
  L. Ding, K. Jin, B. Ying, K. Yuan, and W. Yin
  The International Conference on Machine Learning (ICML)
  [Code]

• AdaNPC: Exploring Non-Parametric Classifier for Test-Time Adaptation
  Y.-F. Zhang, X. Wang, K. Jin, K. Yuan, Z. Zhang, L. Wang, R. Jin, and T. Tan
  The International Conference on Machine Learning (ICML)
  [Code]

• BEVHeight: A Robust Framework for Vision-based Roadside 3D Object Detection
  L. Yang, K. Yu, T. Tang, J. Li, K. Yuan, L. Wang, X. Zhang, and P. Chen
  The IEEE/CVF Computer Vision and Pattern Recognition Conference (CVPR)
  [Code]

2022

• Revisiting optimal convergence rate for smooth and non-convex stochastic decentralized optimization
  K. Yuan, X. Huang, Y. Chen, X. Zhang, Y. Zhang, and P. Pan
  The Conference on Neural Information Processing Systems (NeurIPS)
  

• Communication-efficient topologies for decentralized learning with O(1) consensus rate
  Z. Song, W. Li, K. Jin, L. Shi, M. Yan, W. Yin, and K. Yuan
  The Conference on Neural Information Processing Systems (NeurIPS)
  [Code] [Poster] [5-min video presentation]

• Lower bounds and nearly optimal algorithms in distributed learning with communication compression
  X. Huang, Y. Chen, W. Yin, and K. Yuan
  The Conference on Neural Information Processing Systems (NeurIPS)

• A unified and refined convergence analysis for non-convex decentralized learning
  S. A. Alghunaim and K. Yuan
  IEEE Transactions on Signal Processing

• A Byzantine-resilient dual subgradient method for vertical federated learning
  K. Yuan, Z. Wu, and Q. Ling
  The IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

• CHEX: Channel exploration for CNN model compression
  Z. Hou, M. Qin, F. Sun, X. Ma, K. Yuan, Y. Xu, Y.-K. Chen, R. Jin, Y. Xie, and S.-Y. Kung
  The IEEE/CVF Computer Vision and Pattern Recognition Conference (CVPR)
  [Code]

• Effective model sparsification by scheduled Grow-and-Prune methods
  X. Ma, M. Qin, F. Sun, Z. Hou, K. Yuan, Y. Xu, Y. Wang, Y.-K. Chen, R. Jin, and Y. Xie
  The International Conference on Learning Representations (ICLR)
  [Code]

2021

• Exponential graph is provably efficient for decentralized deep training
  B. Ying, K. Yuan, Y. Chen, H. Hu, P. Pan, and W. Yin
  The Conference on Neural Information Processing Systems (NeurIPS)
  [Code]

• Improved analysis and rates for variance reduction under without-replacement sampling orders
  X. Huang, K. Yuan, X. Mao, and W. Yin
  The Conference on Neural Information Processing Systems (NeurIPS)
  

• DecentLaM: Decentralized momentum SGD for large-batch deep training
  K. Yuan, Y. Chen, X. Huang, Y. Zhang, P. Pan, Y. Xu, and W. Yin
  The International Conference on Computer Vision (ICCV)

• Accelerating gossip SGD with periodic global averaging
  Y. Chen, K. Yuan, Y. Zhang, P. Pan, Y. Xu, and W. Yin
  The International Conference on Machine Learning (ICML)

2020

• Multiagent fully decentralized value function learning with linear convergence rates
  L. Cassano, K. Yuan, and A. H. Sayed
  IEEE Transactions on Automatic Control   Short version appeared in ECC 2019

• Decentralized proximal gradient algorithms with linear convergence rates
  S. A. Alghunaim, E. K. Ryu, K. Yuan, and A. H. Sayed
  IEEE Transactions on Automatic Control

• Walkman: A communication-efficient random-walk algorithm for decentralized optimization
  X. Mao, K. Yuan, Y. Hu, Y. Gu, and A. H. Sayed, W. Yin
  IEEE Transactions on Signal Processing

• Can primal methods outperform primal-dual methods in decentralized dynamic optimization?
  K. Yuan, W. Xu, and Q. Ling
  IEEE Transactions on Signal Processing   Short version appeared in Asilomar 2019

• On the influence of bias-correction on distributed stochastic optimization
  K. Yuan, S. A. Alghunaim, B. Ying, and A. H. Sayed
  IEEE Transactions on Signal Processing   Short version appeared in CDC 2019

• Variance-reduced stochastic learning under random reshuffling
  B. Ying, K. Yuan, and A. H. Sayed
  IEEE Transactions on Signal Processing   Short version appeared in ICASSP 2018

2019

• A proximal diffusion strategy for multiagent optimization with sparse affine constraints
  S. A. Alghunaim, K. Yuan, and A. H. Sayed
  IEEE Transactions on Automatic Control

• Dynamic average diffusion with randomized coordinate updates
  B. Ying, K. Yuan, and A. H. Sayed
  IEEE Transactions on Signal and Information Processing over Networks

• Decentralized dynamic admm with quantized and censored communications
  Y. Liu, K. Yuan, G. Wu, Z. Tian, and Q. Ling
  Asilomar Conference on Signals, Systems, and Computers

• COVER: A cluster-based variance reduced method for online learning
  K. Yuan, B. Ying, and A. H. Sayed
  The International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

• A linearly convergent proximal gradient algorithm for decentralized optimization
  S. A. Alghunaim, K. Yuan, and A. H. Sayed
  The Conference on Neural Information Processing Systems (NeurIPS)

2018

• Dual coupled diffusion for distributed optimization with affine constraints
  S. A. Alghunaim, K. Yuan, and A. H. Sayed
  The IEEE Conference on Decision and Control (CDC)

• Supervised learning under distributed features
  B. Ying, K. Yuan, and A. H. Sayed
  IEEE Transactions on Signal Processing   Short version appeared in IEEE Data Science Workshop (DSW) 2018

• Stochastic learning under random reshuffling with constant step-sizes
  B. Ying, K. Yuan, S. Vlaski, and A. H. Sayed
  IEEE Transactions on Signal Processing

• Exact diffusion for distributed optimization and learning – Part I: Algorithm development
  K. Yuan, B. Ying, X. Zhao, and A. H. Sayed
  IEEE Transactions on Signal Processing   Short version appeared in EUSIPCO 2017

• Exact diffusion for distributed optimization and learning – Part II: Convergence analysis
  K. Yuan, B. Ying, X. Zhao, and A. H. Sayed
  IEEE Transactions on Signal Processing

• Variance-reduced stochastic learning by networked agents under random reshuffling
  K. Yuan, B. Ying, J. Liu, and A. H. Sayed
  IEEE Transactions on Signal Processing   Short version appeared in EUSIPCO 2018

2017

• Decentralized exact coupled optimization
  S. A. Alghunaim, K. Yuan, and A. H. Sayed
  Allerton Conference on Communication, Control, and Computing (Allerton)

• Decentralized consensus optimization with asynchrony and delays
  T. Wu, K. Yuan, Q. Ling, W. Yin, and A. H. Sayed
  IEEE Transactions on Signal and Information Processing over Networks

• On the performance of random reshuffling in stochastic learning
  B. Ying, K. Yuan, S. Vlaski, and A. H. Sayed
  Information Theory and Application Workshop (ITA)

2016

• Stochastic gradient descent with finite samples sizes
  K. Yuan, B. Ying, S. Vlaski, and A. H. Sayed
  IEEE Workshop on Machine Learning for Signal Processing (MLSP)

• Online dual coordinate ascent learning
  B. Ying, K. Yuan, and A. H. Sayed
  European Signal Processing Conference (EUSIPCO)

• On the influence of momentum acceleration on online learning
  K. Yuan, B. Ying, and A. H. Sayed
  Journal of Machine Learning Research

• On the convergence of decentralized gradient descent
  K. Yuan, Q. Ling, and W. Yin
  SIAM Journal on Optimization
  ICCM Distinguished Paper Award
  Highly Cited Paper in SIAM Journal on Optimization

2015

• A decentralised linear programming approach to energy–efficient event detection
  K. Yuan, Q. Ling, and Z. Tian
  International Journal of Sensor Networks

2014

• Communication-efficient decentralized event monitoring in wireless sensor networks
  K. Yuan, Q. Ling, and Z. Tian
  IEEE Transactions on Parallel and Distributed Systems

• On the linear convergence of the ADMM in decentralized consensus optimization
  W. Shi, Q. Ling, K. Yuan, G, Wu, and W. Yin
  IEEE Transactions on Signal Processing
  IEEE Signal Processing Society Yougn Author Best Paper Award [Awardee list] [News]
  Highly Cited Paper in IEEE Transactions on Signal Processing
  

2013

• A linearized Bregman algorithm for decentralized basis pursuit
  K. Yuan, Q. Ling, W. Yin, and A. Ribeiro
  European Signal Processing Conference (EUSIPCO)

• Linearly convergent decentralized consensus optimization with the alternating direction method of multipliers
  W. Shi, Q. Ling, K. Yuan, G, Wu, and W. Yin
  The IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)