arXiv subjects

Hiroki Kuji

Publications and source records attributed to Hiroki Kuji.

8 recordsLinked to original sources

arXiv2026

Experimental Measurement and Theoretical Analysis of Energy Relaxation Rates of an Interacting Two-Qubit System on a D-Wave Quantum Annealer

In D-Wave quantum annealers, various properties of the ground state have been clarified, whereas the correspondence between theory and experiment for energy relaxation rates in multiqubit excited states remains insufficiently understood. Here, we measured the energy relaxation rates of the first excited states in single-qubit systems and interacting two-qubit systems using a D-Wave quantum annealer. We analyzed the measured relaxation rates using a Gorini-Kossakowski-Sudarshan-Lindblad master equation with independent local $σ^{\,\,z}$-type noise channels. The calculated relaxation rates reproduced the overall trends observed experimentally, supporting the model at a qualitative level. We then used the relaxation rates measured for the uncoupled single-qubit systems to calibrate the local relaxation parameters and predict the relaxation rates of the interacting two-qubit systems. The predicted rates remained within a factor of approximately four of the measured rates. These results show that the relaxation measurements on individual qubits can provide a practical estimate of relaxation in small interacting quantum systems and may help clarify relaxation mechanisms in programmable quantum annealers.

quant-ph↗

arXiv2026

Quantum Error Mitigation Simulates General Non-Hermitian Dynamics

While non-Hermitian Hamiltonians enable exotic dynamical phenomena, implementing their nonunitary time evolution on near-term quantum devices remains challenging. We propose a hardware-friendly protocol that simulates non-Hermitian dynamics without ancillas, controlled time evolution, or continuous monitoring. The protocol combines a Gorini-Kossakowski-Sudarshan-Lindblad (GKSL) evolution via classical Gaussian white-noise averaging with stochastic quantum error mitigation (QEM) to cancel the jump contribution at the level of expectation values. The mitigation layer uses only single-qubit operations. We validate the method through numerical simulations of an asymmetric-hopping model and an open {\it XXZ} spin chain with non-Hermitian boundary fields. Our work provides a programmable and ancilla-free framework for investigating exotic dynamics beyond the class of completely positive and trace-preserving dynamics using QEM.

quant-ph↗

arXiv2026

Characterization and generation of a SQL-beating catlike state through repetitive measurements

Sensitivity in metrology without entanglement is limited by the standard quantum limit (SQL). Recent studies have found that the Heisenberg-limited scaling, the ultimate sensitivity in quantum metrology, can be achieved by generalized cat states, which are characterized by an index that indicates coherence among macroscopically distinct states and are associated with additive observables. Although generalized cat states include diverse states, encompassing classical mixtures of exponentially large numbers of states, the preparation of large generalized cat states has not been demonstrated yet. Here we characterize SQL-beating catlike states using the index $q$ indicating macroscopic coherence and prove that any state with $q>1.5$ has a potential to surpass the SQL when used as a sensor. We propose a protocol to generate them through repetitive measurements on a quantum spin system of $N$ spins, which we call a spin ensemble. Starting from a thermal equilibrium state of the spin ensemble, we demonstrate that we can increase the coherence among the spin ensemble via repetitive weak measurements of its total magnetization, which is indirectly measured through an ancillary qubit collectively coupled to the ensemble. Notably, our method for creating the SQL-beating catlike states requires no dynamical control over the spin ensemble. As a potential experimental realization, we discuss a hybrid system composed of a superconducting flux qubit and donor spins in silicon. Our results pave the way for the realization of entanglement-enhanced quantum metrology in state-of-the-art technology.

quant-ph↗

arXiv2025

Variational quantum-neural hybrid imaginary time evolution

Numerous methodologies have been proposed to implement imaginary time evolution (ITE) on quantum computers. Among these, variational ITE (VITE) methods for noisy intermediate-scale quantum (NISQ) computers have attracted much attention, which uses parametrized quantum circuits to mimic non-unitary dynamics. Although widely studied, conventional variational quantum algorithms including face challenges in achieving high accuracy due to their strong dependence on the choice of ansatz quantum circuits. Recently, the variational quantum-neural hybrid eigensolver (VQNHE), which combines the neural network (NN) with a variational quantum eigensolver, has been proposed. This approach enhances the expressive power of variational states and improves the estimation of expectation values. Motivated by this idea, we explore the hybridization of VITE with a NN-based non-unitary operator. In this study, we propose a method named variational quantum-neural hybrid ITE (VQNHITE). By combining the NN and parameterized quantum circuit, our proposal enhances the expressive power compared to conventional approaches, enabling more accurate tracking of imaginary-time dynamics. In addition, to mitigate the instability arising from randomly initialized NN parameters, we introduce an initial-parameter optimization procedure at a small imaginary-time step, which stabilizes the subsequent variational evolution. We tested our approach with numerical simulations on Heisenberg spin chains under both nearest-neighbor and all-to-all circuit connectivities. The results demonstrate that VQNHITE consistently achieves higher fidelity with the exact ITE state compared to VITE.

quant-ph↗

arXiv2025

Proposal for realizing quantum-spin systems on a two-dimensional square lattice with Dzyaloshinskii-Moriya interaction by Floquet engineering using Rydberg atoms

We theoretically propose a method for implementing the Hamiltonian incorporating Heisenberg and Dzyaloshinskii-Moriya (DM) interactions within Rydberg atoms arranged in a two-dimensional square lattice, utilizing Floquet engineering. In our scheme, we use both global and local operations of the spins. The global operations can be realized by applying the microwave and the local operations can be realized by the locally addressing lasers, which yields the ac-Stark shift. Since our engineered Hamiltonian contains bond-dependent DM interactions, we expect the emergence of quantum skyrmions in the ground state.

cond-mat.quant-gas↗

arXiv2025

Robust phase estimation of the ground-state energy without controlled time evolution on a quantum device

Estimating the ground-state energy of Hamiltonians in quantum systems is an important task. In this work, we demonstrate that the ground-state energy can be accurately estimated without controlled time evolution by using adiabatic state preparation (ASP) and Ramsey-type measurement. By considering the symmetry of the Hamiltonian governing the time evolution during ASP, we can prepare a superposition of the ground state and reference state whose eigenvalue is known. This enables the estimation of the ground-state energy via Ramsey-type measurement. Furthermore, our method is robust against non-adiabatic transitions, making it suitable for use with early fault-tolerant quantum computers and quantum annealing.

quant-ph↗

arXiv2025

Quantum Circuit Learning Using Non-Integrable System Dynamics

Quantum machine learning is an approach that aims to improve the performance of machine learning methods by leveraging the properties of quantum computers. In quantum circuit learning (QCL), a supervised learning method that can be implemented using variational quantum algorithms (VQAs), the process of encoding input data into quantum states has been widely discussed for its important role on the expressive power of learning models. In particular, the properties of the eigenvalues of the Hamiltonian used for encoding significantly influence model performance. Recent encoding methods have demonstrated that the expressive power of learning models can be enhanced by applying exponentially large magnetic fields proportional to the number of qubits. However, this approach poses a challenge as it requires exponentially increasing magnetic fields, which are impractical for implementation in large-scale systems. Here, we propose a QCL method that leverages a non-integrable Hamiltonian for encoding, aiming to achieve both enhanced expressive power and practical feasibility. We find that the thermalization properties of non-integrable systems over long timescales, implying that the energy difference has a low probability to be degenerate, lead to an enhanced expressive power for QCL. Since the required magnetic field strength remains within a practical range, our approach to using the non-integrable system is suitable for large-scale quantum computers. Our results bridge the dynamics of non-integrable systems and the field of quantum machine learning, suggesting the potential for significant interdisciplinary contributions.

quant-ph↗

arXiv2025

Identification of Phase Plate Properties Using Photonic Quantum Sensor Networks

Quantum sensor networks (QSNs) have been widely studied for their potential of precise measurements. While most QSN research has focused on estimating continuous variables, recent studies have explored discrete-variable estimation. Here, we propose a method for high-precision identification of phase plate properties using a photon-based QSN, which is categorized as discrete-variable estimation. We consider an interaction of a single photon with $N$ phase plates. There are some distinct properties of the phase plates, and we aim to identify such properties. Specifically, we investigate two cases: (i) distinguishing between phase plates that impart uniformly random phases in the range $[0, 2π]$ and those that impart the same phase, and (ii) distinguishing between phase plates that impart uniformly random phases in $[0, 2π]$ and those that impart phases within a narrower range $[- δ, δ]$ ($0< δ\ll 1$). For this distinction, we consider two approaches: one in which a single photon is prepared in a nonlocal state before interacting with the phase plates, and the other in which the single photon remains in a local state. Our results demonstrate that the nonlocal state enables more precise identification when $N$ is large.

quant-ph↗