Scholar
Liang Zhang
Google Scholar ID: OIgmMCkAAAAJ
ETH Zurich
Citations & Impact
Citations
161
H-index
7
i10-index
6
Publications
15
Co-authors
12
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Publications include: DPZero [ICML'24], Zeroth-Order Optimization Finds Flat Minima [NeurIPS'25], PoLAR: Polar-Decomposed Low-Rank Adapter Representation [NeurIPS'25], On the Crucial Role of Initialization for Matrix Factorization [ICLR'25], Implicit Regularization of Sharpness-Aware Minimization for Scale-Invariant Problems [NeurIPS'24], etc.
Research Experience
- Closely collaborates with Sewoong Oh (University of Washington) and Kiran Koshy Thekumparampil (Amazon).
Education
- Ph.D. in Computer Science, Max Planck ETH Center for Learning Systems, 2021-Present, Advisors: Niao He (ETH Zurich) and Michael Muehlebach (Max Planck Institute for Intelligent Systems); M.S. in Computer Science, ETH Zurich, 2021; B.S. in Physics, Peking University, 2019.
Background
- Research Interests: Intersection between optimization and machine learning, particularly in developing efficient optimization algorithms for trustworthy machine learning. Current research targets efficient fine-tuning of large language models (LLMs).
Miscellany
- Presented at AI+X Summit 2025 in Zurich on efficient LLMs fine-tuning (ELF).
