Center Image

Zhen Qin
Ohio State University
Department of Computer Science and Engineering

Email: qin.660@osu.edu

Zhen Qin

(I am seeking a postdoctoral position starting in autumn or winter 2025 and am open to opportunities across various fields. Please feel free to contact me if you have relevant openings.)

Hello, I am Zhen Qin, a fourth-year PhD candidate in the Department of Computer Science and Engineering at Ohio State University. I am privileged to be under the guidance of Prof. Zhihui Zhu. My current research encompasses several intriguing domains:

Structured optimization and theoretical foundations for tensor learning, signal processing, communication, quantum tomography and machine learning,
Special structures of high-dimensional space including sparse, low-rank, manifold-based and tensor networks-based models.

Research Overview: My research focuses on the design and analysis of models and algorithms in scientific and engineering domains, leveraging numerical optimization, mathematical signal processing, and information theory. In particular, I study the learning theory of large-scale models with structured properties in quantum information, communication systems, and neural network architectures. By deepening theoretical insights, I aim to further enhance and refine practical models, bridging the gap between theory and real-world applications. alt text

News and Updates

[Jan 2025] Our paper has been released at arXiv, presenting a projected classical shadow method for quantum state tomography.
[Dec 2024] Our paper has been accepted at JMLR, offering a convergence guarantee for the factorization approach in arbitrary-order tensor train recovery.
[Dec 2024] Our paper has been released at arXiv, providing optimal error analysis of channel estimation for IRS-assisted MIMO systems.
[Nov 2024] Our paper has been released at arXiv, investigating the optimal allocation of Pauli measurements in the low-rank quantum state tomography.
[Oct 2024] Our paper has been released at arXiv, offering a theoretical analysis of the robust tensor train (TT) recovery problem and demonstrating that TT-format tensors can be robustly recovered even when up to half of the measurements are arbitrarily corrupted.
[Oct 2024] Our paper has been released at arXiv, proving that a linear number of state copies is required to guarantee bounded recovery error of an matrix product operator state in the quantum state tomography, thereby improving the theoretical result in our TIT paper.
[Jun 2024] Our paper has been released at arXiv, offering computational and statistical guarantees for tensor-on-tensor regression with tensor train decomposition.
[Mar 2024] Our paper has been accepted at SPL, analyzing the linear converence rate of training the orthonormal deep linear neural networks.
[Jan 2024] Our paper has been accepted at TIT, demonstrating that a polynomial number of state copies is required to guarantee bounded recovery error of an matrix product operator state in the quantum state tomography.
[Jan 2024] A series of proportionate recursive least squares (PRLS) algorithms have been completed and accepted in the following papers: paper: PRLS, paper: L1-PRLS, paper: VSS-CR-PRLS and paper: Fast PRLS, exploring proportionate sparsity in the adaptive signal processing during my master's studies.

Select Publications

Z. Qin and Z. Zhu, ‘‘ Robust Low-rank Tensor Train Recovery”, arXiv preprint arXiv:2410.15224, 2024.
Z. Qin, C. Jameson, A. Goldar, Z. Gong, M. B. Wakin and Z. Zhu, ‘‘ Sample-Optimal Quantum State Tomography for Structured Quantum States in One Dimension”, arXiv preprint arXiv.2410.02583, 2024.
Z. Qin and Z. Zhu, ‘‘ Computational and Statistical Guarantees for Tensor-on-Tensor Regression with Tensor Train Decomposition”, arXiv preprint arXiv:2406.06002, 2024.
Z. Qin, C. Jameson, Z. Gong, M. B. Wakin and Z. Zhu, ‘‘ Quantum State Tomography for Matrix Product Density Operators”, IEEE Transactions on Information Theory (TIT), 2024.
Z. Qin, M. B. Wakin and Z. Zhu, ‘‘ Guaranteed Nonconvex Factorization Approach for Tensor Train Recovery”, Journal of Machine Learning Research (JMLR), 2024.