王铭泽

王铭泽 (Mingze Wang) 博士候选人 北京大学 数学科学学院 北京大学 静园六院 210 室 北京大学 20 楼 210 室 中国 北京, 100084 电子邮箱: mingzewang [at] stu [dot] pku [dot] edu [dot] cn [Google Scholar]     [简历]

关于我

我是北京大学数学科学学院计算数学专业的四年级直博生 (2021-现在)。 我非常荣幸能得到鄂维南院士的指导。 在此之前，我于 2021 年在浙江大学数学科学学院获得了数学与应用数学学士学位 (本科前三年排名为 1/111)。

研究兴趣

我对机器学习的理论、算法和应用有着广泛的兴趣。我对非凸和凸优化也很感兴趣。

最近，我致力于使用理论来优雅地设计算法。

具体来说，我最近的研究课题是

• 深度学习理论：优化理论、泛化理论、隐式偏好、表达能力。[1][2][3][4][5][6][8][9][10][11][12]

• Transformer 和大型语言模型：理论与算法。[8][10][12]

• 非凸和凸优化：理论与算法。[2][4][6][10][11][12]

• CV 和 NLP：算法与应用。[7]

发表论文

• [10] Improving Generalization and Convergence by Enhancing Implicit Regularization
  Mingze Wang, Jinbo Wang, Haotian He, Zilin Wang, Guanhua Huang, Feiyu Xiong, Zhiyu Li, Weinan E, Lei Wu
  2024 Conference on Neural Information Processing Systems (NeurIPS 2024), 1-35.

• [9] Loss Symmetry and Noise Equilibrium of Stochastic Gradient Descent
  Liu Ziyin, Mingze Wang, Hongchao Li, Lei Wu
  2024 Conference on Neural Information Processing Systems (NeurIPS 2024), 1-26.

• [8] Understanding the Expressive Power and Mechanisms of Transformer for Sequence Modeling
  Mingze Wang, Weinan E
  2024 Conference on Neural Information Processing Systems (NeurIPS 2024), 1-70.

• [7] Are AI-Generated Text Detectors Robust to Adversarial Perturbations?
  Guanhua Huang, Yuchen Zhang, Zhe Li, Yongjian You, Mingze Wang, Zhouwang Yang
  2024 Annual Meeting of the Association for Computational Linguistics (ACL 2024), 1-20.

• [6] Achieving Margin Maximization Exponentially Fast via Progressive Norm Rescaling
  Mingze Wang, Zeping Min, Lei Wu
  2024 International Conference on Machine Learning (ICML 2024), 1-38.

• [5] A Theoretical Analysis of Noise Geometry in Stochastic Gradient Descent
  Mingze Wang, Lei Wu
  NeurIPS 2023 Workshop on Mathematics of Modern Machine Learning (NeurIPS 2023 - M3L), 1-30.
  

• [4] Understanding Multi-phase Optimization Dynamics and Rich Nonlinear Behaviors of ReLU Networks
  Mingze Wang, Chao Ma
  2023 Conference on Neural Information Processing Systems (NeurIPS 2023) (Spotlight, 前 3.5%), 1-94.

• [3] The alignment property of SGD noise and how it helps select flat minima: A stability analysis
  Lei Wu, Mingze Wang, Weijie J. Su
  2022 Conference on Neural Information Processing Systems (NeurIPS 2022), 1-25.

• [2] Early Stage Convergence and Global Convergence of Training Mildly Parameterized Neural Networks
  Mingze Wang, Chao Ma
  2022 Conference on Neural Information Processing Systems (NeurIPS 2022), 1-73.

预印本论文

* 表示平等贡献.

• [12] How Transformers Implement Induction Heads: Approximation and Optimization Analysis
  Mingze Wang*, Ruoxi Yu*, Weinan E, Lei Wu.
  arXiv preprint, 1-39. Oct 2024.

• [11] Sharpness-Aware Minimization Efficiently Selects Flatter Minima Late in Training
  Zhanpeng Zhou*, Mingze Wang*, Yuchen Mao, Bingrui Li, Junchi Yan.
  arXiv preprint, 1-24. Oct 2024.

• [1] Generalization Error Bounds for Deep Neural Networks Trained by SGD
  Mingze Wang, Chao Ma
  arXiv preprint, 1-32, June 2022.

部分奖项及荣誉

• 国家奖学金 (博士) (前 2%), 教育部, 2024.

• 校长奖学金, 北京大学, 2024.

• 北大数学研究生奖 (前 1%), 北京大学, 2023.

• 北京大学学术创新奖 (前 1%), 北京大学, 2022.

• 浙江省优秀毕业生 (前 5%), 浙江省, 2021.

• 浙江大学一等奖学金 (前 3%), 浙江大学, 2019; 2020.

• 国家奖学金 (本科) (前 2%), 教育部, 2019.