Research

Research Interests

My current interests include

• Second-Order Inertial Dynamical System and Applications to Optimization.

• Optimal Transport.

• Minimax Optimization with Applications to Game Theory.

• Large-Scale Text Mining.

Refereed Publications

• A Unified Adaptive Tensor Approximation Scheme to Accelerate Composite Convex Optimization

  • (α-β order) Bo Jiang, Tianyi Lin and Shuzhong Zhang

  • To appear in SIAM Journal on Optimization, 2020

• Structured Nonconvex Optimization Models: Algorithms and Iteration Complexity Analysis

  • (α-β order) Bo Jiang, Tianyi Lin, Shiqian Ma and Shuzhong Zhang

  • Computational Optimization and Applications, 72 (1): 115-157, 2019

• On the Iteration Complexity Analysis of Stochastic Primal-Dual Hybrid Gradient approach with High Probability

  • Linbo Qiao, Tianyi Lin, Qi Qin and Xicheng Lu

  • Neurocomputing, 307: 78-90, 2018

• 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, 76 (1): 69-88, 2018

• Stochastic Primal-Dual Proximal ExtraGradient Descent for Compositely Regularized Optimization

  • Tianyi Lin, Linbo Qiao, Teng Zhang, Jiashi Feng and Bofeng Zhang

  • Neurocomputing, 273: 516-525, 2018

• Distributed Linearized Alternating Direction Method of Multipliers for Composite Convex Consensus Optimization

  • Necdet S. Aybat, Zi Wang, Tianyi Lin and Shiqian Ma

  • IEEE Transactions on Automatic Control, 63 (1): 5-20, 2018

• An Extragradient-Based Alternating Direction Method for Convex Minimization

  • Tianyi Lin, Shiqian Ma and Shuzhong Zhang

  • Foundations of Computational Mathematics, 17 (1): 35-59, 2017

• 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, 52 (1): 221-254, 2017

• 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, 69: 52-81, 2016

• On the Sublinear Convergence Rate of Multi-Block ADMM

  • Tianyi Lin, Shiqian Ma and Shuzhong Zhang

  • Journal of the Operations Research Society of China, 3(3): 251-274, 2015

• On the Global Linear Convergence of the ADMM with Multi-Block Variables

  • Tianyi Lin, Shiqian Ma and Shuzhong Zhang

  • SIAM Journal on Optimization, 25 (3): 1478-1497, 2015

Refereed Proceedings

• New Proximal Newton-Type Methods for Convex Optimization

  • (α-β order) Ilan Adler, Zhiyue T. Hu and Tianyi Lin

  • CDC’20, Jeju Island, Korea

• Finite-Time Last-Iterate Convergence for Multi-Agent Learning in Games

  • Tianyi Lin, Zhengyuan Zhou (equal contribution), Panayotis Mertikopoulos and Michael I. Jordan

  • ICML’20, Vienna, Austria

• On Gradient Descent Ascent for Nonconvex-Concave Minimax Problems

  • Tianyi Lin, Chi Jin and Michael I. Jordan

  • ICML’20, Vienna, Austria

• Near-Optimal Algorithms for Minimax Optimization

  • Tianyi Lin, Chi Jin and Michael I. Jordan

  • COLT’20, Graz, Austria

• Improved Sample Complexity for Stochastic Compositional Variance Reduced Gradient

  • Tianyi Lin, Chenyou Fan, Mengdi Wang and Michael I. Jordan

  • ACC’20, Denver, USA

• On Efficient Optimal Transport: An Analysis of Greedy and Accelerated Mirror Descent Algorithms

  • Tianyi Lin, Nhat Ho (equal contribution) and Michael I. Jordan

  • ICML’19, Long Beach, USA

• Sparsemax and Relaxed Wasserstein for Topic Sparsity

  • Tianyi Lin, Zhiyue T. Hu and Xin Guo

  • WSDM’19, Melbourne, Australia

• Understanding Sparse Topical Structure of Short Text via Stochastic Variational-Gibbs Inference

  • Tianyi Lin, Siyuan Zhang and Hong Cheng

  • CIKM’16, Indianapolis, USA

• On Stochastic Primal-Dual Hybrid Gradient Approach for Compositely Regularized Minimization

  • Linbo Qiao, Tianyi Lin, Yugang Jiang, Fan Yang, Wei Liu and Xicheng Lu

  • ECAI’16, the Hague, Netherlands

• Collaborative Filtering Incorporating Review Text and Co-clusters of Hidden User Communities and Item Groups

  • Yinqing Xu, Wai Lam and Tianyi Lin

  • CIKM’14, Shanghai, China

• Latent Aspect Mining via Exploring Sparsity and Intrinsic Information

  • Yinqing Xu, Tianyi Lin, Wai Lam, Zirui Zhou, Hong Cheng and Anthony M-C. So,

  • CIKM’14, Shanghai, China

• The Dual-Sparse Topic Model: Mining Focused Topics and Focused Terms in Short Text

  • Tianyi Lin, Wentao Tian (equal contribution), Qiaozhu Mei and Hong Cheng

  • WWW’14, Seoul, Korea

Preprints

• On Projection Robust Optimal Transport: Sample Complexity and Model Misspecification

  • Tianyi Lin, Zeyu Zheng, Elynn Y. Chen, Marco Cuturi, Michael I. Jordan

• Projection Robust Wasserstein Distance and Riemannian Optimization

  • Tianyi Lin, Chenyou Fan (equal contribution), Nhat Ho, Macro Cuturi and Michael I. Jordan

• Fixed-Support Wasserstein Barycenters: Computational Hardness and Fast Algorithm

  • Tianyi Lin, Nhat Ho, Xi Chen, Macro Cuturi and Michael I. Jordan

• A Control-Theoretic Perspective on Optimal High-Order Optimization

  • Tianyi Lin and Michael I. Jordan

• On the Complexity of Approximating Multimarginal Optimal Transport

  • Tianyi Lin, Nhat Ho, Macro Cuturi and Michael I. Jordan

• On the Efficiency of the Sinkhorn and Greenkhorn Algorithms and Their Acceleration for Optimal Transport

  • Tianyi Lin, Nhat Ho and Michael I. Jordan

• On Structured Filtering-Clustering: Global Error Bound and Optimal First-Order Algorithms

  • Nhat Ho, Tianyi Lin (equal contribution) and Michael I. Jordan

• An ADMM-Based Interior-Point Method for Large-Scale Linear Programming

  • Tianyi Lin, Shiqian Ma, Yinyu Ye and Shuzhong Zhang

• Adaptively Accelerating Cubic Regularized Newton's Methods for Convex Optimization via Random Sampling

  • (α-β order) Xi Chen, Bo Jiang, Tianyi Lin and Shuzhong Zhang