Wei Liu

News

• 09.11 New preprint: A sequential smoothing majorant stochastic approximation method for nonconvex nonconcave minimax problems.
• 06.27 New GRF grant funded by the Hong Kong RGC: Oracle complexity bounds of first-order methods for nonsmooth optimization with nonconvex function constraints.
• 03.14 Paper accepted by Transactions on Machine Learning Research: “LoDAdaC: a unified local training-based decentralized framework with Adam-type updates and compressed communication”.
• 01.26 Paper accepted by SIAM Journal on Optimization: A SPIDER-type stochastic subgradient method for expectation-constrained nonconvex nonsmooth optimization.
• 01.06 Paper published in Mathematical Programming Computation: Damped proximal augmented Lagrangian method for weakly-convex problems with convex constraints.
• 07.09 Paper accepted by Transactions on Machine Learning Research: Compressed decentralized momentum stochastic gradient methods for nonconvex optimization.

About me

I received my bachelor’s degree in mathematics from Zhejiang University in 2017. In September 2017, I began my master’s studies at the Academy of Mathematics and Systems Science of the Chinese Academy of Sciences, under the supervision of Professor Xin Liu, and became a Ph.D. candidate in August 2019. From August 2019 to August 2021, I visited The Hong Kong Polytechnic University, hosted by Professor Xiaojun Chen. I obtained my Ph.D. in June 2022.

From 2022 to 2025, I was a Postdoctoral Research Associate at Rensselaer Polytechnic Institute, working with Professor Yangyang Xu. I then joined the Department of Applied Mathematics at The Hong Kong Polytechnic University as a Research Assistant Professor. I have joined Wuhan University in 2027 as a Research Fellow and tenure-track Associate Professor.

Research interests

My research focuses on the design and analysis of (stochastic) first-order methods for nonlinear optimization. I aim to develop algorithms that are both simple and efficient. A central idea is to make difficult problems more tractable through exact penalization, smoothing, and convexification.

These transformations allow us to use well-established algorithmic frameworks, with strong theoretical guarantees and practical performance. My interests include:

• First-order methods for large-scale optimization.
• Stochastic models and methods for statistical data analysis and machine learning.
• Complexity analysis: iteration, oracle, and computational complexity.
• Applications in AI and machine learning, including fairness, deep neural networks, distributed robust optimization, decentralized distributed learning, semi-supervised learning, bilevel optimization, minimax problems, and large language models.

More about my research · Get in touch

Selected publications

• Wei Liu and Yangyang Xu. A SPIDER-type stochastic subgradient method for expectation-constrained nonconvex nonsmooth optimization. SIAM Journal on Optimization, 2026. [arXiv] [Code]
• Wei Liu, Muhammad Khan, Gabriel Mancino-Ball, and Yangyang Xu. A stochastic smoothing framework for nonconvex-nonconcave min-expectation-max problems with applications to Wasserstein distributionally robust optimization. [arXiv]
• Hari Dahal, Wei Liu, and Yangyang Xu. Damped proximal augmented Lagrangian method for weakly-convex problems with convex constraints. Mathematical Programming Computation, 2026. [arXiv] [MPC]
• Wei Liu, Qihang Lin, and Yangyang Xu. First-order methods for affinely constrained composite non-convex non-smooth problems: Lower complexity bound and near-optimal methods. [arXiv] [Related MOR paper]
• Wei Liu, Xin Liu, and Xiaojun Chen. An inexact augmented Lagrangian algorithm for training leaky ReLU neural network with group sparsity. Journal of Machine Learning Research, 2023. [arXiv] [Code] [JMLR]
• Wei Liu, Xin Liu, and Xiaojun Chen. Linearly-constrained nonsmooth optimization for training autoencoders. SIAM Journal on Optimization, 2022. [arXiv] [Code] [SIOPT]

All publications and preprints →

Education

• 2017–2022 · Institute of Computational Mathematics and Scientific/Engineering Computing, Academy of Mathematics and Systems Science, China.
  Ph.D. in Computational Mathematics. Advisor: Xin Liu.
• 2013–2017 · Zhejiang University, China.
  B.S. in Mathematics, Mathematics Pursuit Science Class, Chu Kochen Honors College (CKC College).

Experiences

• August 2019–August 2021 · The Hong Kong Polytechnic University.
  Research Assistant, hosted by Professor Xiaojun Chen, Chair Professor of Applied Mathematics.
• May 2023 · Brown University.
  Workshop Visiting Scholar.
• August 2022–August 2025 · Rensselaer Polytechnic Institute.
  Postdoctoral Research Associate, hosted by Professor Yangyang Xu.

Selected awards

• 2026 · Selected for the National High-Level Young Talent Program.
• 2022 · CAS President’s Scholarship.
• 2017 · Outstanding Graduate, Zhejiang University (First Prize).
• 2014–2016 · Basic Disciplines Top-notch Student Scholarship (First Prize), awarded three times.

Skills

• Programming languages: MATLAB, C, Python.
• Languages: Chinese (native); English (professional working proficiency).
• Professional knowledge: Nonsmooth analysis, convex optimization, machine learning, and first-order methods in optimization.

Journal refereeing

1. Mathematics of Operations Research
2. Mathematical Programming Computation
3. Computational Optimization and Applications
4. Journal of Machine Learning Research
5. SCIENCE CHINA Mathematics
6. IEEE Transactions on Cybernetics
7. Journal of Computational Mathematics
8. Journal of Scientific Computing