中文简介
尹超，河海大学数学学院讲师。2019 年本科毕业于北京理工大学数学与统计学院，2024 年直博毕业于南京大学数学系，导师为杨俊锋教授。 研究兴趣包括分布式优化、联邦学习、双层规划以及一阶算法理论。 近期工作围绕去中心化分布式优化、双层优化、自适应步长与原始-对偶算法，关注通信受限、数据异质和大规模机器学习场景下的收敛性分析与高效算法设计。 主持国家自然科学基金青年项目和江苏省自然科学基金青年项目。
河海大学数学学院 School of Mathematics, Hohai University
尹超 Chao Yin
讲师，数学学院 Lecturer, School of Mathematics
我是河海大学数学学院讲师，主要研究分布式优化、联邦学习、双层规划以及一阶算法理论。 近期工作关注去中心化分布式优化、双层优化、自适应步长算法和原始-对偶方法。
I am a lecturer at the School of Mathematics, Hohai University. My research focuses on distributed optimization, federated learning, bilevel optimization, and first-order algorithms.
About
个人简介 Biography
English Bio
Chao Yin is a lecturer at the School of Mathematics, Hohai University. He received his B.S. degree from the School of Mathematics and Statistics, Beijing Institute of Technology in 2019, and his Ph.D. degree from the Department of Mathematics, Nanjing University in 2024, under the supervision of Prof. Junfeng Yang. His research focuses on distributed optimization, federated learning, first-order methods, and bilevel optimization. His recent work studies decentralized bilevel optimization, adaptive stepsizes, primal-dual splitting, ADMM-type methods, and convergence theory under communication constraints and data heterogeneity. He leads youth projects funded by the National Natural Science Foundation of China and the Natural Science Foundation of Jiangsu Province.
Research
研究方向 Research Interests
一阶算法 First-Order Methods
一阶优化算法、收敛性分析、复杂度界和参数选择机制。
First-order optimization algorithms, convergence analysis, complexity bounds, and parameter selection.
分布式优化 Distributed Optimization
去中心化算法、梯度跟踪、通信约束与网络结构影响。
Decentralized algorithms, gradient tracking, communication limits, and graph effects.
联邦学习 Federated Learning
自适应方法、异质数据、客户端采样和鲁棒训练。
Adaptive methods, data heterogeneity, client sampling, and robust training.
双层规划 Bilevel Optimization
双层优化、超参数优化、数据清洗和元学习中的算法设计与收敛分析。
Algorithm design and convergence analysis for bilevel optimization, hyperparameter optimization, data cleaning, and meta-learning.
Publications
论文与预印本 Publications and Preprints
精选论文 Selected Publications
2026
S3LDBO: A Snapshot Single-Loop Algorithm for Decentralized Bilevel Optimization
Chao Yin, Youran Dong, Shiqian Ma, Bofan Wang, and Junfeng Yang. arXiv 预印本，2026。 arXiv preprint, 2026.
2025
2024
A Modified Primal-Dual Algorithm for Structured Convex Optimization with a Lipschitzian Term
Chao Yin, Hai-Wen Xu, and Jun-Feng Yang. Journal of the Operations Research Society of China, 2024。 Journal of the Operations Research Society of China, 2024.
2024
Golden Ratio Proximal Gradient ADMM for Distributed Composite Convex Optimization
Chao Yin and Junfeng Yang. Journal of Optimization Theory and Applications, 200(3):895-922, 2024。 Journal of Optimization Theory and Applications, 200(3):895-922, 2024.
Grants
基金项目 Funded Projects
弱光滑条件下的去中心分布式一阶算法研究及其应用 Decentralized First-Order Algorithms under Weak Smoothness and Applications
国家自然科学基金青年项目，2026.01.01-至今，经费 30 万，主持
NSFC Young Scientists Fund, 2026.01.01-present, RMB 300,000, PI
采取自适应步长的去中心分布式算法研究及其应用 Decentralized Distributed Algorithms with Adaptive Stepsizes and Applications
江苏省自然科学基金青年项目，2025.09.01-至今，经费 20 万，主持
Natural Science Foundation of Jiangsu Province Young Scholars Program, 2025.09.01-present, RMB 200,000, PI
Teaching & Recruiting
教学与招生 Teaching and Student Recruiting
招生方向 Student Recruiting
- 欢迎对优化算法与理论、大模型训练与使用感兴趣的学生联系。
- Students interested in optimization algorithms, theory, and LLM training or usage are welcome to contact me.
能力要求 Expected Skills
- 熟练掌握 Python 和 LaTeX，具备较好的数学基础与论文阅读能力。
- Proficiency in Python and LaTeX, with solid mathematical training and paper-reading ability.
- 鼓励熟练使用 Codex、Claude Code 等 AI 工具开展科研与编程工作。
- Experience with AI research and coding tools such as Codex and Claude Code is encouraged.
Personal
个人爱好 Outside Research
运动与摄影 Sports and Photography
- 喜欢篮球，羽毛球，乒乓球，台球，游泳等运动娱乐。
- I enjoy basketball for its teamwork, rhythm, and split-second decision making.
- 喜欢摄影，尤其关注旅行、风景和日常生活中的光线与构图。
- I also enjoy photography, especially travel, landscapes, and quiet moments of everyday life.
游戏与技术 Games and Technology
- 业余时间也喜欢游戏，常把它看作放松、策略思考和体验交互设计的一种方式。
- In my spare time, I like games as a way to relax, think strategically, and appreciate interactive design.
- 我也持续关注 AI 工具如何改变科研写作、代码实现和数学实验的日常工作流。
- I am also interested in how AI tools reshape research writing, coding, and mathematical experimentation.
Contact
联系方式 Contact
邮箱： Email: yinchao@hhu.edu.cn
单位： Affiliation: 河海大学数学学院 School of Mathematics, Hohai University
办公室： Office: 江宁校区励学楼 B301 Room B301, Lixue Building, Jiangning Campus
ORCID: 0009-0007-3070-0824
ResearchGate: Chao Yin