I am a fifth-year PhD student in Computer Sciences at the University of Wisconsin–Madison. I am broadly interested in optimization and computational learning theory, with a primary focus on understanding the hidden structures in foundational machine learning models that make gradient-descent-like algorithms effective. I am very fortunate to be advised by Prof. Jelena Diakonikolas. I have also had the pleasure of collaborating with Prof. Ilias Diakonikolas and many other amazing collaborators. Before coming to Madison, I earned my B.S. in Mathematics from Shandong University.

Publication

• ARO: A New Lens On Matrix Optimization For Large Models

  Wenbo Gong, Javier Zazo, Qijun Luo, Puqian Wang, James Hensman, Chao Ma, In submission

• Robustly Learning Monotone Single-Index Models

  Puqian Wang*, Nikos Zarifis*, Ilias Diakonikolas, Jelena Diakonikolas, NeurIPS 2025 arxiv

  (* Equal Contribution)

• Robustly Learning Monotone Generalized Linear Models via Data Augmentation

  Nikos Zarifis*, Puqian Wang*, Ilias Diakonikolas, Jelena Diakonikolas, COLT 2025 arxiv

  (* Equal Contribution)

• Sample and Computationally Efficient Robust Learning of Gaussian Single-Index Models

  Puqian Wang, Nikos Zarifis, Ilias Diakonikolas, Jelena Diakonikolas, NeurIPS 2024, arxiv

• Robustly Learning Single-Index Models via Alignment Sharpness

  Nikos Zarifis*, Puqian Wang*, Ilias Diakonikolas, Jelena Diakonikolas, ICML 2024, arxiv

  (* Equal Contribution)

• Near-Optimal Bounds for Learning Gaussian Halfspaces with Random Classification Noise

  ($\alpha\beta$) Ilias Diakonikolas, Jelena Diakonikolas, Daniel M Kane, Puqian Wang, Nikos Zarifis, NeurIPS, 2023, arxiv

• Information-Computation Tradeoffs for Learning Margin Halfspces with Random Classification Noise

  ($\alpha\beta$) Ilias Diakonikolas, Jelena Diakonikolas, Daniel M Kane, Puqian Wang, Nikos Zarifis, COLT, 2023, arxiv

• Robustly Learning a Single Neuron via Sharpness

  Puqian Wang* , Nikos Zarifis* , Ilias Diakonikolas, Jelena Diakonikolas, ICML, 2023, Oral Presentation, arxiv

  (* Equal Contribution)

• Potential Function-based Framework for Making the Gradients Small in Convex and Min-Max Optimization

  Jelena Diakonikolas, Puqian Wang, SIAM Journal on Optimization, 2022, arxiv

Talks

• Sample and Computationally Efficient Robust Learning of Gaussian Single-Index Models, IFDS 2024 Meeting, October 2024, UW Madison
• Robustly Learning Single-Index Models via Alignment Sharpness, 60th Annual Allerton Conference on Communication, Control, and Computing, September 2024, Urbana-Champaign, IL, US
• Robustly Learning Single-Index Models via Alignment Sharpness, International Symposium on Mathematical Programming, July 2024, Montreal, Q.C., Cananda

Experience

Research Intern at Microsoft Research, Cambridge, UK, Summer 2025.

Teaching

TA@UW Madison

• Fall 2021 CS577 Introduction to Algorithms
• Fall 2022 CS726 Nonlinear Optimization I