Publications

2024

  1. Perseus: A simple and optimal high-order method for variational inequalities
    Tianyi Lin, and Michael I Jordan
    Mathematical Programming (Series A), 2024
    arXiv HTML
  2. Doubly optimal no-regret online learning in strongly monotone games with bandit feedback
    Wenjia Ba, Tianyi Lin, Jiawei Zhang, and Zhengyuan Zhou
    Operations Research, 2024
    arXiv HTML
  3. Adaptive, doubly optimal no-regret learning in strongly monotone and exp-concave games with gradient feedback
    Michael I Jordan, Tianyi Lin, and Zhengyuan Zhou
    Operations Research, 2024
    arXiv HTML
  4. A continuous-time perspective on global acceleration for monotone equation problems
    Tianyi Lin, and Michael I Jordan
    Communications in Optimization Theory, (Invited paper on Special issue dedicated to the memory of Professor Hedy Attouch) , 2024
    arXiv HTML
  5. A specialized semismooth Newton method for kernel-based optimal transport
    Tianyi Lin, Marco Cuturi, and Michael I Jordan
    In International Conference on Artificial Intelligence and Statistics (AISTATS) , 2024
    arXiv HTML

2023

  1. Curvature-independent last-iterate convergence for games on Riemannian manifolds
    Yang Cai, Michael I Jordan, Tianyi Lin, Argyris Oikonomou, and Emmanouil-Vasileios Vlatakis-Gkaragkounis
    ArXiv Preprint, 2023
    arXiv
  2. Structure-driven algorithm design in optimization and machine learning
    Tianyi Lin
    UC Berkeley , 2023
    HTML
  3. Monotone inclusions, acceleration, and closed-loop control
    Tianyi Lin, and Michael I Jordan
    Mathematics of Operations Research, 2023
    arXiv HTML
  4. First-order algorithms for nonlinear generalized Nash equilibrium problems
    Michael I Jordan, Tianyi Lin, and Manolis Zampetakis
    Journal of Machine Learning Research, 2023
    arXiv HTML
  5. Deterministic nonsmooth nonconvex optimization
    Michael I Jordan, Guy Kornowski, Tianyi Lin, Ohad Shamir, and Manolis Zampetakis
    In Conference on Learning Theory (COLT) , 2023
    arXiv HTML

2022

  1. Two-timescale gradient descent ascent algorithms for nonconvex minimax optimization
    Tianyi Lin, Chi Jin, and Michael I Jordan
    Preprint, 2022
    arXiv HTML
  2. Explicit second-order min-max optimization methods with optimal convergence guarantee
    Tianyi Lin, Panayotis Mertikopoulos, and Michael I Jordan
    Preprint, 2022
    arXiv
  3. A control-theoretic perspective on optimal high-order optimization
    Tianyi Lin, and Michael I Jordan
    Mathematical Programming (Series A), 2022
    arXiv HTML
  4. On the efficiency of entropic regularized algorithms for optimal transport
    Tianyi Lin, Nhat Ho, and Michael I Jordan
    Journal of Machine Learning Research, 2022
    arXiv HTML
  5. Accelerating adaptive cubic regularization of Newton’s method via random sampling
    Xi Chen, Bo Jiang, Tianyi Lin, and Shuzhong Zhang
    Journal of Machine Learning Research, 2022
    arXiv HTML
  6. On the complexity of approximating multimarginal optimal transport
    Tianyi Lin, Nhat Ho, Marco Cuturi, and Michael I Jordan
    Journal of Machine Learning Research, 2022
    arXiv HTML
  7. Gradient-free methods for deterministic and stochastic nonsmooth nonconvex optimization
    Tianyi Lin, Zeyu Zheng, and Michael Jordan
    In International Conference on Neural Information Processing Systems (NeurIPS) , 2022
    arXiv HTML
  8. First-order algorithms for min-max optimization in geodesic metric spaces
    Michael I Jordan, Tianyi Lin, and Emmanouil-Vasileios Vlatakis-Gkaragkounis
    In International Conference on Neural Information Processing Systems (NeurIPS) , 2022
    arXiv HTML
  9. Online nonsubmodular minimization with delayed costs: From full information to bandit feedback
    Tianyi Lin, Aldo Pacchiano, Yaodong Yu, and Michael I Jordan
    In International Conference on Machine Learning (ICML) , 2022
    arXiv HTML
  10. Fast distributionally robust learning with variance-reduced min-max optimization
    Yaodong Yu, Tianyi Lin, Eric V Mazumdar, and Michael I Jordan
    In International Conference on Artificial Intelligence and Statistics (AISTATS) , 2022
    arXiv HTML
  11. On structured filtering-clustering: Global error bound and optimal first-order algorithms
    Nhat Ho, Tianyi Lin, and Michael I Jordan
    In International Conference on Artificial Intelligence and Statistics (AISTATS) , 2022
    arXiv HTML

2021

  1. An ADMM-based interior-point method for large-scale linear programming
    Tianyi Lin, Shiqian Ma, Yinyu Ye, and Shuzhong Zhang
    Optimization Methods and Software, (Invited paper on Special issue dedicated to the memory of Professor Masao Iri) , 2021
    arXiv HTML
  2. A variational inequality approach to Bayesian regression games
    Wenshuo Guo, Michael I Jordan, and Tianyi Lin
    In IEEE Conference on Decision and Control (CDC) , 2021
    arXiv HTML
  3. Relaxed Wasserstein with applications to GANs
    Xin Guo, Johnny Hong, Tianyi Lin, and Nan Yang
    In International Conference on Acoustics, Speech and Signal Processing (ICASSP) , 2021
    arXiv HTML
  4. On projection robust optimal transport: Sample complexity and model misspecification
    Tianyi Lin, Zeyu Zheng, Elynn Chen, Marco Cuturi, and Michael I Jordan
    In International Conference on Artificial Intelligence and Statistics (AISTATS) , 2021
    arXiv HTML

2020

  1. A unified adaptive tensor approximation scheme to accelerate composite convex optimization
    Bo Jiang, Tianyi Lin, and Shuzhong Zhang
    SIAM Journal on Optimization, 2020
    arXiv HTML
  2. Fixed-support Wasserstein barycenters: Computational hardness and fast algorithm
    Tianyi Lin, Nhat Ho, Xi Chen, Marco Cuturi, and Michael I Jordan
    In International Conference on Neural Information Processing Systems (NeurIPS) , 2020
    arXiv HTML
  3. Projection robust Wasserstein distance and Riemannian optimization
    Tianyi Lin, Chenyou Fan, Nhat Ho, Marco Cuturi, and Michael I Jordan
    In International Conference on Neural Information Processing Systems (NeurIPS) , 2020
    arXiv HTML
  4. New proximal Newton-type methods for convex optimization
    Ilan Adler, Zhiyue T Hu, and Tianyi Lin
    In IEEE Conference on Decision and Control (CDC) , 2020
    arXiv HTML
  5. On gradient descent ascent for nonconvex-concave minimax problems
    Tianyi Lin, Chi Jin, and Michael I Jordan
    In International Conference on Machine Learning (ICML) , 2020
    arXiv HTML
  6. Finite-time last-iterate convergence for multi-agent learning in games
    Tianyi Lin, Zhengyuan Zhou, Panayotis Mertikopoulos, and Michael I Jordan
    In International Conference on Machine Learning (ICML) , 2020
    arXiv HTML
  7. Near-optimal algorithms for minimax optimization
    Tianyi Lin, Chi Jin, and Michael I Jordan
    In Conference on Learning Theory (COLT) , 2020
    arXiv HTML
  8. Improved sample complexity for stochastic compositional variance reduced gradient
    Tianyi Lin, Chenyou Fan, Mengdi Wang, and Michael I Jordan
    In American Control Conference (ACC) , 2020
    arXiv HTML

2019

  1. Structured nonconvex and nonsmooth optimization: Algorithms and iteration complexity analysis
    Bo Jiang, Tianyi Lin, Shiqian Ma, and Shuzhong Zhang
    Computational Optimization and Applications, 2019
    arXiv HTML
  2. On efficient optimal transport: An analysis of greedy and accelerated mirror descent algorithms
    Tianyi Lin, Nhat Ho, and Michael Jordan
    In International Conference on Machine Learning (ICML) , 2019
    arXiv HTML
  3. Sparsemax and relaxed Wasserstein for topic sparsity
    Tianyi Lin, Zhiyue Hu, and Xin Guo
    In International Conference on Web Search and Data Mining (WSDM) , 2019
    arXiv HTML

2018

  1. Global convergence of unmodified 3-block ADMM for a class of convex minimization problems
    Tianyi Lin, Shiqian Ma, and Shuzhong Zhang
    Journal of Scientific Computing, 2018
    arXiv HTML
  2. On the iteration complexity analysis of stochastic pimal-dual hybrid gradient approach with high probability
    Linbo Qiao, Tianyi Lin, Qi Qin, and Xicheng Lu
    Neurocomputing, 2018
    arXiv HTML
  3. Stochastic primal-dual proximal extragradient descent for compositely regularized optimization
    Tianyi Lin, Linbo Qiao, Teng Zhang, Jiashi Feng, and Bofeng Zhang
    Neurocomputing, 2018
    arXiv HTML

2017

  1. Distributed linearized alternating direction method of multipliers for composite convex consensus optimization
    Necdet Serhat Aybat, Zi Wang, Tianyi Lin, and Shiqian Ma
    IEEE Transactions on Automatic Control, 2017
    arXiv HTML
  2. An extragradient-based alternating direction method for convex minimization
    Tianyi Lin, Shiqian Ma, and Shuzhong Zhang
    Foundations of Computational Mathematics, 2017
    arXiv HTML
  3. Exploiting interactions of review text, hidden user communities and item groups, and time for collaborative filtering
    Yinqing Xu, Qian Yu, Wai Lam, and Tianyi Lin
    Knowledge and Information Systems, 2017
    HTML

2016

  1. Iteration complexity analysis of multi-block ADMM for a family of convex minimization without strong convexity
    Tianyi Lin, Shiqian Ma, and Shuzhong Zhang
    Journal of Scientific Computing, 2016
    arXiv HTML
  2. Understanding sparse topical structure of short text via stochastic variational-Gibbs inference
    Tianyi Lin, Siyuan Zhang, and Hong Cheng
    In International on Conference on Information and Knowledge Management (CIKM) , 2016
    HTML
  3. On stochastic primal-dual hybrid gradient approach for compositely regularized minimization
    Linbo Qiao, Tianyi Lin, Yu-Gang Jiang, Fan Yang, Wei Liu, and Xicheng Lu
    In European Conference on Artificial Intelligence (ECAI) , 2016
    HTML

2015

  1. On the global linear convergence of the ADMM with multiblock variables
    Tianyi Lin, Shiqian Ma, and Shuzhong Zhang
    SIAM Journal on Optimization, 2015
    arXiv HTML
  2. On the sublinear convergence rate of multi-block ADMM
    Tianyi Lin, Shiqian Ma, and Shuzhong Zhang
    Journal of the Operations Research Society of China, 2015
    arXiv HTML

2014

  1. Collaborative filtering incorporating review text and co-clusters of hidden user communities and item groups
    Yinqing Xu, Wai Lam, and Tianyi Lin
    In International Conference on Information and Knowledge Management (CIKM) , 2014
    HTML
  2. Latent aspect mining via exploring sparsity and intrinsic information
    Yinqing Xu, Tianyi Lin, Wai Lam, Zirui Zhou, Hong Cheng, and Anthony Man-Cho So
    In International Conference on Information and Knowledge Management (CIKM) , 2014
    HTML
  3. The dual-sparse topic model: mining focused topics and focused terms in short text
    Tianyi Lin, Wentao Tian, Qiaozhu Mei, and Hong Cheng
    In International Conference on World Wide Web (WWW) , 2014
    HTML