Biography

Education

Ph.D. in Statistics, 09/2024 - present
School of Management, University of Science and Technology of China (USTC)
Advisor: Prof. Xueqin Wang
Research: Optimization algorithms for large-scale machine learning, distributed optimization, best subset selection.

M.S. in Data Science, 09/2021 - 06/2024
School of Artificial Intelligence and Data Science, University of Science and Technology of China (USTC)
Advisor: Prof. Xueqin Wang
Research: Distributed best subset selection algorithms, semismooth Newton methods.

Distributed Best Subset Selection, 2022 - present

Proposed a two-stage distributed algorithm achieving minimax optimal rate with oracle property under \(\ell_0\) constraints.
Developed a data-driven Generalized Information Criterion for adaptive sparsity selection.
Result: arXiv:2408.17276

Communication-Efficient Decentralized Optimization, 2024 - present

Co-developed a double-communication Symmetric ADMM framework for decentralized networks.
Achieved net reduction in total communication cost via multi-round communication per iteration.
Result: arXiv:2511.05283

Distributed Semismooth Newton Method, 2023 - present

Designed a dual-framework semismooth Newton solver for distributed sparse optimization.
Achieved linear convergence with only one vector transmission per iteration.
Demonstrated significant speed advantages over first-order methods and ADMM on Lasso and quantile regression.

Coordinate-wise ADMM (Co-ADMM), 2022 - present

Proposed a coordinate-wise ADMM for piecewise linear/quadratic + convex penalty optimization.
Achieved linear convergence with \(O(n)\) per-iteration complexity.
Developed a solver outperforming CVX, ECOS, and SCS on SVM, Huber regression, and quantile regression.

Teaching Assistant at USTC:

Python, R, C++, Julia, Matlab