Xu Zhiqian

Email: xuzhiqian20z@ict.ac.cn

Education

Ph.D. in Computer Software and Theory

Institute of Computing Technology, Chinese Academy of Sciences & University of Chinese Academy of Sciences
Sep 2020 - Jun 2025
Location: Beijing

  • Laboratory: Parallel Software Research Group, Research Center for High-Performance Computing (Supervisor: Prof. Honghui Shang & Prof. Yunquan Zhang)
  • Research Interests: Parallel Computing, Quantum Chemistry, Tensor Networks
  • Honors: Excellent Award of the Institute of Computing Technology (2023), Outstanding Student Award of UCAS (2024)

B.Sc. in Computer Science and Technology

School of the Gifted Young, University of Science and Technology of China (Innovation Pilot Class)
Sep 2016 - Jun 2020
Location: Hefei

  • GPA: 3.5 / 4.3

Research Experience

High-Performance Variational Quantum Monte Carlo Solver Based on Matrix Product States

Aug 2023 - Present

  • Proposed a variational quantum Monte Carlo algorithm based on matrix product states and designed an efficient sampling algorithm using binary/quadtree autoregressive sampling.
  • Implemented local energy calculations based on the Slater-Condon rule, time-dependent variational principle algorithms, and non-random configuration selection algorithms in variational quantum Monte Carlo to accelerate convergence.
  • Leveraged PyTorch framework for network inference and GPU acceleration with automatic differentiation.

Scalable and Differentiable Quantum Chemistry Simulator Based on Matrix Product States

Aug 2022 - Aug 2023

  • Addressed the memory consumption issue of automatic differentiation in tensor network-based quantum simulators by designing a quantum circuit backpropagation algorithm with low memory consumption, requiring at most two quantum state copies during the forward process, independent of the number of quantum circuit parameters.
  • Proposed a method to partition quantum chemistry Hamiltonians to parallelize the backpropagation process, achieving 87%/99% strong/weak scalability efficiency with 100 qubits and 3000 processes.

Large-Scale Scalable Quantum Chemistry Simulator Based on Matrix Product States

Nov 2021 - Aug 2022

  • Employed matrix product states as the backend for quantum circuit simulators in variational quantum eigensolvers (VQE) to overcome the exponential memory bottleneck of state vector simulators. Designed a numerically stable single/double quantum gate evolution process, allowing the simulator to simulate deeper circuits.
  • Developed a parallel circuit evolution algorithm based on VQE quantum circuit characteristics, scaling to 20 million cores on the Sunway supercomputer with 92%/91% strong/weak scalability efficiency.

Publications

  1. MPS-VQE: A Variational Quantum Computational Chemistry Simulator with Matrix Product States, Computer Physics Communications (SCI Q2, First Author)
  2. Scalable and Differentiable Simulator for Quantum Computational Chemistry, IPDPS’24 (CCF Category B Conference, First Author)
  3. Differentiable Matrix Product States for Simulating Variational Quantum Computational Chemistry, Quantum (SCI Q2, Third Author)
  4. Large-scale Simulation of Quantum Computational Chemistry on a New Sunway Supercomputer, SC’22 (CCF Category A Conference, Fourth Author)

Skills

  • Programming Languages: C/C++, Python, Julia, Fortran
  • Professional Skills: Parallel Programming (MPI, OpenMP), Heterogeneous Optimization (Sunway Supercomputer, GPU)
  • English Proficiency: Passed CET-6