Jianhao Ma is a tenure-track Assistant Professor in the Department of Industrial Engineering at Tsinghua University. He received his Ph.D. in Industrial and Operations Engineering from the University of Michigan, Ann Arbor, in 2025, where he was advised by Professor Salar Fattahi. From 2025 to 2026, he was a postdoctoral researcher in the Department of Statistics and Data Science at the University of Pennsylvania, working with Professor Yuxin Chen. He received a B.E. in Industrial Engineering and a B.S. in Mathematics from Tsinghua University and was an exchange student in the Department of Statistics at the University of California, Berkeley.

His research focuses on optimization for machine learning and model compression. He develops theory and algorithms for efficient and reliable machine learning, with current interests spanning nonconvex optimization and implicit regularization, robust low-rank recovery, model quantization, and optimization for large-model training.

· July 2026–Present: Tenure-track Assistant Professor, Department of Industrial Engineering, Tsinghua University

· June 2025–June 2026: Postdoctoral Researcher, Department of Statistics and Data Science, University of Pennsylvania

· January 2021–May 2025: Ph.D. in Industrial and Operations Engineering, University of Michigan, Ann Arbor; advisor: Professor Salar Fattahi

· September 2016–June 2020: B.E. in Industrial Engineering and B.S. in Mathematics, Tsinghua University

· January 2019–August 2019: Exchange Student, Department of Statistics, University of California, Berkeley

· Optimization theory and algorithms for machine learning

· Nonconvex optimization, optimization trajectories, and implicit regularization

· Model compression, quantization, and quantization-aware training

· Robust machine learning and robust matrix/subspace recovery

· Optimizers and training strategies for large models

· ProQuest Distinguished Dissertation Award, University of Michigan, 2026

· Rackham Predoctoral Fellowship, University of Michigan, 2024–2025

· Second Place, INFORMS Junior Faculty Interest Group Paper Competition (as a coauthor), 2023

· Katta Murty Prize for Best Research Paper on Optimization, Department of Industrial and Operations Engineering, University of Michigan, 2023

· NeurIPS Scholar Award, 2022

Recent Preprints

1. Jianhao Ma and Yuxin Chen. “WSqD: A Horizon-Free Learning Rate Schedule for Large Model Training.” Paper

2. Tianqi Shen, Jinji Yang, Runze Shi, Jianhao Ma, Jiaye Teng, and Ziye Ma. “Towards Understanding the Power and Limits of the Muon Optimizer: A River-Valley Perspective.” Paper

3. Hanyang Li, Jianhao Ma, and Ying Cui. “Understanding Quantization-Aware Training: Gradients at Quantized Weights Bias to the Low-Loss Basin.” Paper

4. Jianhao Ma, Yu Huang, Yuejie Chi, and Yuxin Chen. “Preconditioning Benefits of Spectral Orthogonalization in Muon.” Paper

Published Papers

1. Jianhao Ma, Rui Ray Chen, Yinghui He, Salar Fattahi, and Wei Hu. “Sparse Mean Estimation in Adversarial Settings via Incremental Learning.” Transactions on Machine Learning Research (TMLR), 2026. Paper

2. Jianhao Ma and Salar Fattahi. “Can Learning Be Explained By Local Optimality In Low-rank Matrix Recovery?” Mathematics of Operations Research, 2025. Paper

3. Lisa Jin, Jianhao Ma, Zechun Liu, Andrey Gromov, Aaron Defazio, and Lin Xiao. “PARQ: Piecewise-Affine Regularized Quantization.” International Conference on Machine Learning (ICML), 2025. Paper

4. Jianhao Ma and Salar Fattahi. “Convergence of Gradient Descent with Small Initialization for Unregularized Matrix Completion.” Conference on Learning Theory (COLT), 2024. Paper

5. Jianhao Ma and Salar Fattahi. “Global Convergence of Sub-gradient Method for Robust Matrix Recovery: Small Initialization, Noisy Measurements, and Over-parameterization.” Journal of Machine Learning Research (JMLR), 2023. Paper

6. Jianhao Ma, Lingjun Guo, and Salar Fattahi. “Behind the Scenes of Gradient Descent: A Trajectory Analysis via Basis Function Decomposition.” International Conference on Learning Representations (ICLR), 2023. Paper

7. Jianhao Ma and Salar Fattahi. “Blessing of Nonconvexity in Deep Linear Models: Depth Flattens the Optimization Landscape Around the True Solution.” Advances in Neural Information Processing Systems (NeurIPS), 2022 (Spotlight). Paper

· May 2024–August 2024: Research Scientist Intern, Meta FAIR Labs; host: Dr. Lin Xiao

· April 2020–July 2020: Machine Learning Engineer Intern, ByteDance AI Lab

· Session Chair, INFORMS Annual Meeting, 2021, 2022, and 2024

· Journal reviewer for IEEE Transactions on Information Theory, IEEE Transactions on Signal Processing, SIAM Journal on Optimization, SIAM Review, Journal of Machine Learning Research, and others

· Conference reviewer for ICML, NeurIPS, ICLR, AISTATS, and others