I am a fourth-year PhD student in the H. Milton Stewart School of Industrial and Systems Engineering (ISyE) at Georgia Tech, advised by Prof. Tuo Zhao. My research focuses on developing principled methodologies and theoretical understanding in machine learning, especially deep learning. I am interested in language models and diffusion models, with an emphasis on function approximation theory and statistical guarantees in learning applications.

Before joining Georgia Tech, I earned a B.S. in Statistics from University of Science and Technology of China (USTC).

Preprints

• Diffusion Model for Manifold Data: Score Decomposition, Curvature and Statistical Complexity (Manuscript upon request)
  Zixuan Zhang, Kaixuan Huang, Mengdi Wang, Tuo Zhao and Minshuo Chen.

• A Minimalist Example of Edge-of-Stability and Progressive Sharpening
  Liming Liu, Zixuan Zhang, Simon Du, Tuo Zhao

• COSMOS: A Hybrid Adaptive Optimizer for Memory-Efficient Training of LLMs
  Liming Liu, Zhenghao Xu, Zixuan Zhang, Hao Kang, Zichong Li, Chen Liang, Weizhu Chen, Tuo Zhao

• LLMs Can Generate a Better Answer by Aggregating Their Own Responses
  Zichong Li, Xinyu Feng, Yuheng Cai, Zixuan Zhang, Tianyi Liu, Chen Liang, Weizhu Chen, Haoyu Wang, Tuo Zhao

Publications

• Nonparametric Classification on Low Dimensional Manifolds using Overparameterized Convolutional Residual Networks
  Zixuan Zhang*, Kaiqi Zhang*, Minshuo Chen, Yuma Takeda, Mengdi Wang, Tuo Zhao, Yu-Xiang Wang
  Annual Conference on Neural Information Processing Systems (NeurIPS), 2024

• Robust Reinforcement Learning from Corrupted Human Feedback
  Alexander Bukharin, Ilgee Hong, Haoming Jiang, Zichong Li, Qingru Zhang, Zixuan Zhang, Tuo Zhao
  Annual Conference on Neural Information Processing Systems (NeurIPS), 2024

• Effective Minkowski dimension of deep nonparametric regression: function approximation and statistical theories
  Zixuan Zhang, Minshuo Chen, Mengdi Wang, Wenjing Liao, Tuo Zhao
  International Conference on Machine Learning (ICML), 2023

• Sequential information design: Markov persuasion process and its efficient reinforcement learning
  Jibang Wu, Zixuan Zhang, Zhe Feng, Zhaoran Wang, Zhuoran Yang, Michael I Jordan, Haifeng Xu
  ACM Conference on Economics and Computation, 2022

Talks

• Sample Complexity of Diffusion Models for Learning Distributions on Low Dimensional Manifolds
  INFORMS Annual 2024, Seattle