A5036638784- Quantum Computing Algorithms and Architecture
- Quantum Information and Cryptography
- Quantum many-body systems
- Quantum and electron transport phenomena
- Quantum Mechanics and Applications
- Quantum Mechanics and Non-Hermitian Physics
- Quantum, superfluid, helium dynamics
- Quantum-Dot Cellular Automata
- Quantum chaos and dynamical systems
- Advanced Thermodynamics and Statistical Mechanics
- Diamond and Carbon-based Materials Research
- 2D Materials and Applications
- Machine Learning in Materials Science
- Perovskite Materials and Applications
- Graphene research and applications
- Atomic and Subatomic Physics Research
- Advanced Memory and Neural Computing
- Neural Networks and Applications
Southern University of Science and Technology
2021-2024
Beijing Academy of Quantum Information Sciences
2023
Principal component analysis (PCA) is a widely applied but rather time-consuming tool in machine learning techniques. In 2014, Lloyd, Mohseni, and Rebentrost proposed quantum PCA (qPCA) algorithm [Lloyd, Rebentrost, Nat. Phys. 10, 631 (2014)] that still lacks experimental demonstration due to the challenges preparing multiple state copies implementing phase estimations. Here, we propose new qPCA using hybrid classical-quantum control, where parameterized circuits are optimized with simple...
Quantum computing is expected to provide exponential speedup in machine learning. However, optimizing the data loading process, commonly referred as quantum embedding, maximize classification performance remains a critical challenge. In this work, we propose neural embedding (NQE) technique based on deterministic computation with one qubit (DQC1). Unlike traditional approach, NQE trains network trace distance between states corresponding different categories of classical data. Furthermore,...
Abstract With the rapid growth of controllable qubits in recent years, experimental multipartite entangled states can be created with high fidelity various moderate‐ and large‐scale physical systems. However, characterization entanglement structure remains a formidable challenge, as traditionally it requires exponential number local measurements to realize identification. Machine learning is demonstrated an efficient tool detect underlying for ideal states, but has non‐negligible...
Abstract With the advancement of computing power and algorithms, machine learning has been a powerful tool in numerous applications nowadays. However, hardware limitation classical computers increasing size datasets urge community to explore new techniques for learning. Quantum‐enhanced is such rapidly growing field. It refers quantum algorithms that are implemented computers, which can improve computational speed tasks often promises an exponential speedup. In past few years, development...
Fidelity estimation is an important technique for evaluating prepared quantum states in noisy devices. A recent theoretical work proposed a frugal approach called neural fidelity (NQFE) [X. Zhang et al., Phys. Rev. Lett. 127, 130503 (2021).]. While this requires much smaller number of measurement operators than full state tomography, it uses weight-based floating strategy that predetermines the top global Pauli contribute most to and discrete intervals as predictions. In Letter, we develop...
The accurate determination of the electronic structure strongly correlated materials using first principle methods is paramount importance in condensed matter physics, computational chemistry, and material science. However, due to exponential scaling resources, incorporating such into classical computation frameworks becomes prohibitively expensive. In 2016, Bauer et al. proposed a hybrid quantum-classical approach [B. al., Hybrid materials, Phys. Rev. X 6, 031045...
Quantum state tomography (QST) via local measurements on reduced density matrices (LQST) is a promising approach but becomes impractical for large systems. To tackle this challenge, we developed an efficient quantum method inspired by overlapping [Phys. Rev. Lett. 124, 100401 (2020)PRLTAO0031-900710.1103/PhysRevLett.124.100401], which utilizes parallel (PQST). In contrast to LQST, PQST significantly reduces the number of and offers more robustness against shot noise. Experimentally,...
Non-Hermitian quantum systems have recently attracted considerable attention due to their exotic properties. Though many experimental realizations of non-Hermitian been reported, the non-Hermiticity usually resorts hard-to-control environments and cannot last for too long times. An alternative approach is use simulation with closed system, whereas how simulate Hamiltonian dynamics remains a great challenge. To tackle this problem, we propose protocol which combines dilation method...
Entanglement is a key property in the development of quantum technologies and study many-body simulations. However, entanglement measurement typically requires full-state tomography (FST). Here we present neural network-assisted protocol for measuring equilibrium non-equilibrium states local Hamiltonians. Instead FST, it can learn comprehensive quantities from single-qubit or two-qubit Pauli measurements, such as R\'enyi entropy, partially-transposed (PT) moments, coherence. It also exciting...
Quantum state tomography (QST) via local measurements on reduced density matrices (LQST) is a promising approach but becomes impractical for large systems. To tackle this challenge, we developed an efficient quantum method inspired by overlapping [Phys. Rev. Lett. 124, 100401(2020)], which utilizes parallel (PQST). In contrast to LQST, PQST significantly reduces the number of and offers more robustness against shot noise. Experimentally, demonstrate feasibility in tree-like superconducting...
In the Hamiltonian-based quantum dynamics, to estimate Hamiltonians from measured data is a vital topic. this work, we propose recurrent neural network learn target temporal records of single-qubit measurements, which does not require ground states and only measures observables. It applicable on both time-independent time-dependent can simultaneously capture magnitude sign Hamiltonian parameters. Taking with nearest-neighbor interactions as numerical examples, trained our networks different...
Grain boundaries (GBs) frequently emerge in a CVD-grown large-scale transition metal dichalcogenides monolayer thin film, which affect the electronic and optical properties of material. Photoluminescence (PL) can be easily quenched/enhanced at GBs, are, however, merely investigated relatively large tilt angles (θ>14°) previous research. Here, we experimentally examine PL WS2 GBs with as small few degrees. Contrary to conventional wisdom, find that intensity remains intact by when...
Machine learning is widely applied in various areas due to its advantages pattern recognition, but it severely restricted by the computing power of classic computers. In recent years, with rapid development quantum technology, machine has been verified experimentally many systems, and exhibited great over classical algorithms for certain specific problems. present review, we mainly introduce two typical spin nuclear magnetic resonance nitrogen-vacancy centers diamond, review some...
Non-Hermitian quantum systems have recently attracted considerable attention due to their exotic properties. Though many experimental realizations of non-Hermitian been reported, the non-Hermiticity usually resorts hard-to-control environments and cannot last for too long times. An alternative approach is use simulation with closed system, whereas how simulate Hamiltonian dynamics remains a great challenge. To tackle this problem, we propose protocol which combines dilation method...