Jianhao Ma

Jianhao Ma Ph.D. student at Department of Industrial and Operational Engineering University of Michigan, Ann Arbor, United States Email: jianhao [at] umich [dot] edu [Google Scholar][Curriculem Vitae]

About me

I'm currently a second-year Ph.D. student at Umich. I am fortunate to be advised by Professor Salar Fattahi. My research focus on understanding the pratical success of gradient-based algorithms based on trajectory analysis. I am welcome for potential collaboration. Please feel free to contact me via Email!

Research Interests

• General optimization and generalization in modern machine learning.

• Empirical process and its application.

Preprints and Publications

• Blessing of Nonconvexity in Deep Linear Models: Depth Flattens the Optimization Landscape Around the True Solution [arxiv]
  Jianhao Ma, Salar Fattahi
  

• Global Convergence of Sub-gradient Method for Robust Matrix Recovery: Small Initialization, Noisy Measurements, and Over-parameterization [arxiv]
  Jianhao Ma, Salar Fattahi
  

• Towards Understanding Generalization via Decomposing Excess Risk Dynamics [paper] [arxiv]
  Jiaye Teng*, Jianhao Ma*, Yang Yuan
  International Conference on Learning Representations (ICLR) 2022
  

• Sign-RIP: A Robust Restricted Isometry Property for Low-rank Matrix Recovery [paper][arxiv]
  Jianhao Ma, Salar Fattahi
  NeurIPS Workshop on Optimization for Machine Learning, 2021
  

Conference Services

Reviewer of ICML (2021-2022), NeurIPS (2021-2022), ICLR (2022).