胡潇尹
  • 教育程度:博士研究生

  • 职称:副教授

  • 电话:

  • 邮箱:hxy@szu.edu.cn

  • 地址:汇星楼421

教程程度 博士研究生 职称 副教授
电话 邮箱 hxy@szu.edu.cn
地址 汇星楼421 教育经历 <p>
<div style="margin: 0.6em 0;">
  2016.09–2021.06,中国科学院数学与系统科学研究院,博士。
</div>

<div style="margin: 0.6em 0;">
  2012.09–2016.06,浙江大学竺可桢学院,学士。
</div>
</p>
工作经历 <p>
<div style="margin: 0.6em 0;">
   2025.10–至今,深圳大学,副教授
</div>

<div style="margin: 0.6em 0;">
   2025.01–2025.09,浙大城市学院,副教授
</div>

<div style="margin: 0.6em 0;">
   2021.07–2024.12,浙大城市学院,讲师
</div>
</p>
研究领域 最优化方法;人工智能中的优化问题;非光滑优化;非凸优化
获得荣誉 教学课程 <div style="margin: 0.6em 0;">
  <span style="color: steelblue; font-weight: bold;">🔹 本科生课程:</span>
  <span style="color: black;">《数学建模》、《高等数学A》</span>
</div>
</p>
科研成果 <p>
<span style="color: steelblue; font-weight: bold;">📚 Accepted/Published papers:</span>

<div style="margin: 0.6em 0;">
[8] <strong>Xiaoyin Hu</strong>, Kim-Chuan Toh, Shiwei Wang and Nachuan Xiao*. A minimization approach for minimax optimization with coupled constraints. <em>SIAM Journal on Optimization</em>, to appear.
</div>

<div style="margin: 0.6em 0;">
[7] Nachuan Xiao, Kuangyu Ding*, <strong>Xiaoyin Hu</strong> and Kim-Chuan Toh. (2026). Developing Lagrangian-based methods for nonsmooth nonconvex optimization. <em>Mathematics of Operations Research</em>.
</div>

<div style="margin: 0.6em 0;">
[6] Nachuan Xiao, <strong>Xiaoyin Hu</strong>*, Xin Liu and Kim-Chuan Toh. (2025). CDOpt: a Python package for a class of Riemannian optimization. <em>Mathematical Programming Computation</em>.
</div>

<div style="margin: 0.6em 0;">
[5] <strong>Xiaoyin Hu</strong>, Nachuan Xiao*, Xin Liu and Kim-Chuan Toh. (2025). Learning-rate-free momentum SGD with reshuffling converges in nonsmooth nonconvex optimization. <em>Journal of Scientific Computing</em>, (3), 102.
</div>

<div style="margin: 0.6em 0;">
[4] <strong>Xiaoyin Hu</strong>, Nachuan Xiao*, Xin Liu and Kim-Chuan Toh. (2024). A constraint dissolving approach for nonsmooth optimization over the Stiefel manifold. <em>IMA Journal on Numerical Analysis</em>, 44(6), 3717–3748.
</div>

<div style="margin: 0.6em 0;">
[3] Nachuan Xiao, <strong>Xiaoyin Hu</strong>*, Xin Liu and Kim-Chuan Toh. (2024). Adam-family methods for nonsmooth optimization with convergence guarantees. <em>Journal of Machine Learning Research</em>, 25(48), 1-53.
</div>

<div style="margin: 0.6em 0;">
[2] <strong>Xiaoyin Hu</strong>, Nachuan Xiao*, Xin Liu and Kim-Chuan Toh. (2023). An improved unconstrained approach for bilevel optimization. <em>SIAM Journal on Optimization</em>, 33(4), 2801-2829.
</div>

<div style="margin: 0.6em 0;">
[1] <strong>Xiaoyin Hu</strong>, Jianshu Li, Xiaoya Li and J.C. Cui*. (2021). Carbon emissions abatement allocation based on inverse slack-based model. <em>Journal of the Operation Research Society of China</em>, 9, 475-498.
</div>

<br>
<span style="color: steelblue; font-weight: bold;">📝 Submitted papers:</span>


<div style="margin: 0.6em 0;">
[1] Nachuan Xiao,<strong>Xiaoyin Hu</strong>*, Kim-Chuan Toh. SGD-type Methods with Guaranteed Global Stability in Nonsmooth Nonconvex Optimization.  <em>arXiv preprint arXiv:2307.10053</em>.  Under Review.
</div>


<div style="margin: 0.6em 0;">
[2] Nachuan Xiao, <strong>Xiaoyin Hu</strong>*, Xin Liu, Kim-Chuan Toh. A hybrid subgradient method for nonsmooth nonconvex bilevel optimization. <em>arXiv preprint arXiv:2505.22040.</em>.  Under Review.
</div>

<div style="margin: 0.6em 0;">
[3] <strong>Xiaoyin Hu</strong>, Xin Liu, Kim-Chuan Toh, Nachuan Xiao*. Partial Envelope for Optimization Problem with Nonconvex Constraints. <em>arXiv preprint arXiv:2510.22223.</em>.  Under Review.
</div>
</p>

科研项目 <p>
<span style="color: steelblue; font-weight: bold;">📌 主持项目(共3项):</span>

<div style="margin: 0.6em 0;">
[3] 国家自然科学基金青年科学基金项目,深度学习中的流形优化问题:算法设计与求解软件包的开发,2024.01-2026.12。
</div>

<div style="margin: 0.6em 0;">
[2] 浙江省自然科学基金探索项目,机器学习中的流形优化问题的算法设计及其理论与应用,2023.01-2025.12。
</div>

<div style="margin: 0.6em 0;">
[1] 浙江省教育厅一般项目,正交约束优化问题的算法设计及其在人工智能中的应用,2022.10-2024.10。
</div>

</p>

个人简介

深圳大学数学科学学院副教授,广东省运筹学会女运筹工作者委员会秘书长,主要的研究方向是最优化理论与方法,尤其是针对在人工智能中有重要应用的约束优化和非光滑优化问题设计高效的求解方法。在SIAM J. Optim., Math. Oper. Res.IMA J. Numer. Anal., J. Mach. Learn. Res., Math. Program. Comput. 等国际学术刊物发表学术论文10余篇,开发流形优化软件包CDOpt(https://cdopt.github.io/),下载量达15万余次。 担任SIOPT、MPC、JSC等期刊审稿人,以及美国《数学评论》评论员

个人主页:https://huxiaoyin708.github.io/

欢迎数学扎实、有Python或者Julia代码基础的同学报考硕士研究生。

配备深度学习训练所需要的高性能工作站,搭载AMD Threadripper 7980X处理器、RTX 5090显卡。

教育经历

  • 2016.09–2021.06,中国科学院数学与系统科学研究院,博士。
    2012.09–2016.06,浙江大学竺可桢学院,学士。

工作经历

  • 2025.10–至今,深圳大学,副教授
    2025.01–2025.09,浙大城市学院,副教授
    2021.07–2024.12,浙大城市学院,讲师

研究领域

  • 最优化方法;人工智能中的优化问题;非光滑优化;非凸优化

获得荣誉

教学课程

  • 🔹 本科生课程: 《数学建模》、《高等数学A》

科研成果

  • 📚 Accepted/Published papers:

    [8] Xiaoyin Hu, Kim-Chuan Toh, Shiwei Wang and Nachuan Xiao*. A minimization approach for minimax optimization with coupled constraints. SIAM Journal on Optimization, to appear.
    [7] Nachuan Xiao, Kuangyu Ding*, Xiaoyin Hu and Kim-Chuan Toh. (2026). Developing Lagrangian-based methods for nonsmooth nonconvex optimization. Mathematics of Operations Research.
    [6] Nachuan Xiao, Xiaoyin Hu*, Xin Liu and Kim-Chuan Toh. (2025). CDOpt: a Python package for a class of Riemannian optimization. Mathematical Programming Computation.
    [5] Xiaoyin Hu, Nachuan Xiao*, Xin Liu and Kim-Chuan Toh. (2025). Learning-rate-free momentum SGD with reshuffling converges in nonsmooth nonconvex optimization. Journal of Scientific Computing, (3), 102.
    [4] Xiaoyin Hu, Nachuan Xiao*, Xin Liu and Kim-Chuan Toh. (2024). A constraint dissolving approach for nonsmooth optimization over the Stiefel manifold. IMA Journal on Numerical Analysis, 44(6), 3717–3748.
    [3] Nachuan Xiao, Xiaoyin Hu*, Xin Liu and Kim-Chuan Toh. (2024). Adam-family methods for nonsmooth optimization with convergence guarantees. Journal of Machine Learning Research, 25(48), 1-53.
    [2] Xiaoyin Hu, Nachuan Xiao*, Xin Liu and Kim-Chuan Toh. (2023). An improved unconstrained approach for bilevel optimization. SIAM Journal on Optimization, 33(4), 2801-2829.
    [1] Xiaoyin Hu, Jianshu Li, Xiaoya Li and J.C. Cui*. (2021). Carbon emissions abatement allocation based on inverse slack-based model. Journal of the Operation Research Society of China, 9, 475-498.

    📝 Submitted papers:
    [1] Nachuan Xiao,Xiaoyin Hu*, Kim-Chuan Toh. SGD-type Methods with Guaranteed Global Stability in Nonsmooth Nonconvex Optimization. arXiv preprint arXiv:2307.10053. Under Review.
    [2] Nachuan Xiao, Xiaoyin Hu*, Xin Liu, Kim-Chuan Toh. A hybrid subgradient method for nonsmooth nonconvex bilevel optimization. arXiv preprint arXiv:2505.22040.. Under Review.
    [3] Xiaoyin Hu, Xin Liu, Kim-Chuan Toh, Nachuan Xiao*. Partial Envelope for Optimization Problem with Nonconvex Constraints. arXiv preprint arXiv:2510.22223.. Under Review.

科研项目

  • 📌 主持项目(共3项):

    [3] 国家自然科学基金青年科学基金项目,深度学习中的流形优化问题:算法设计与求解软件包的开发,2024.01-2026.12。
    [2] 浙江省自然科学基金探索项目,机器学习中的流形优化问题的算法设计及其理论与应用,2023.01-2025.12。
    [1] 浙江省教育厅一般项目,正交约束优化问题的算法设计及其在人工智能中的应用,2022.10-2024.10。