arXiv subjects

Hiromichi Matsuyama

Publications and source records attributed to Hiromichi Matsuyama.

16 recordsLinked to original sources

arXiv2026

Transformers as Intrinsic Optimizers for Quantum Approximate Optimization Algorithm

The Quantum Approximate Optimization Algorithm (QAOA) is a leading variational framework for combinatorial optimization on noisy intermediate-scale quantum hardware, but its practical performance depends strongly on the classical optimizer used to train its variational parameters. This outer-loop optimization is often nonconvex, initialization-sensitive, and costly when repeated across large families of related problem instances. In this work, we propose a Transformer-based intrinsic optimization framework for QAOA, in which the optimizer itself is learned and embedded directly into the hybrid quantum-classical loop. The proposed graph-conditioned Transformer processes problem structure, current QAOA parameters, measurement feedback, and recent optimization history to predict the next variational-parameter update, thereby reformulating instance-wise classical optimization as an amortized learned policy. We develop a mathematical formulation of this intrinsic-optimization perspective and evaluate the method on QAOA-based MaxCut benchmarks across multiple problem settings, with comparisons against representative classical and learned optimization baselines. The results demonstrate that Transformer-based intrinsic optimization can provide a structured and transferable mechanism for improving the classical component of hybrid quantum optimization algorithms.

quant-ph↗

arXiv2026

Geometric Quantum Physics Informed Neural Network

Quantum physics-informed neural networks (QPINNs) have recently emerged as a promising framework for the solution of partial differential equations (PDEs), with several studies reporting improved convergence and accuracy relative to classical physics-informed neural networks (PINNs) at reduced training cost. Motivated by these advances, we introduce geometric quantum physics-informed neural networks (GQPINNs), a symmetry-aware extension of QPINNs in which the geometric structure of the underlying PDE is incorporated directly into the quantum-circuit ansatz. Building on the framework of geometric quantum machine learning, we construct parametrized circuits that encode finite-group and compact Lie-group symmetries as inductive biases through problem-specific equivariant generator sets . Using a twirling-based construction, we derive symmetry-preserving gates that ensure that the model predictions respect the symmetries of the governing equation whenever the boundary and initial data are symmetry compatible. We benchmark GQPINNs against standard QPINNs and symmetry-adapted classical PINN baselines under matched training protocols across a representative set of linear and nonlinear PDEs. Across these benchmarks, GQPINNs achieve improved solution accuracy, as quantified by lower mean absolute error, while requiring substantially fewer trainable parameters. These results identify symmetry-aware quantum-circuit design as an effective route toward improved efficiency and generalization in quantum PDE solvers and provide a systematic framework for incorporating geometric inductive biases into quantum-enhanced scientific machine learning.

quant-ph↗

arXiv2026

Recursive QAOA for Interference-Aware Resource Allocation in Wireless Networks

Discrete radio resource management problems in dense wireless networks are naturally cast as quadratic unconstrained binary optimization (QUBO) programs but are difficult to solve at scale. We investigate a quantum-classical approach based on the Recursive Quantum Approximate Optimization Algorithm (RQAOA), which interleaves shallow QAOA layers with variable elimination guided by measured single- and two-qubit correlators. For interference-aware channel assignment, we give a compact QUBO/Ising formulation in which pairwise interference induces same-channel couplings and one-hot constraints are enforced via quadratic penalties (or, optionally, constraint-preserving mixers). Within RQAOA, fixing high-confidence variables or relations reduces the problem dimension, stabilizes training, and concentrates measurement effort on a shrinking instance that is solved exactly once below a cutoff. On simulated instances of modest size, including a four-user, four-channel example, the method consistently returns feasible assignments and, for the demonstrated case, attains the global optimum. These results indicate that recursion can mitigate parameter growth and feasibility issues that affect plain QAOA, and suggest a viable pathway for near-term quantum heuristics in wireless resource allocation.

quant-ph↗

arXiv2025

Quantum Annealing with Qubit-Resonator Systems for Simultaneous Optimization of Binary and Continuous Variables

Quantum annealing is a method developed to solve combinatorial optimization problems by utilizing quantum bits. Solving such problems corresponds to minimizing a cost function defined over binary variables. However, in many practical cases, the cost function may also involve continuous variables. Representing continuous variables using quantum bits requires binary encoding, which demands a large number of qubits. To overcome this limitation, an approach using quantum resonators has been proposed, enabling the direct handling of continuous variables within the quantum annealing framework. On the other hand, certain optimization problems involve both binary and continuous variables simultaneously, and a quantum annealing method capable of efficiently solving such hybrid problems has not been established. Here, we propose a quantum annealing method based on a hybrid system composed of qubits and resonators, aiming to minimize cost functions that contain both binary and continuous variables. We present a general framework for hybrid quantum annealing using such systems, and investigate its feasibility and effectiveness through numerical simulations.

quant-ph↗

arXiv2025

Transferring linearly fixed QAOA angles: performance and real device results

Quantum Approximate Optimization Algorithm (QAOA) enables solving combinatorial optimization problems on quantum computers by optimizing variational parameters for quantum circuits. We investigate a simplified approach that combines linear parameterization with parameter transferring, reducing the parameter space to just 4 dimensions regardless of the number of layers. This simplification draws inspiration from quantum annealing schedules providing both theoretical grounding and practical advantages. We compare this combined approach with standard QAOA and other parameter setting strategies such as INTERP and FOURIER, which require computationally demanding incremental layer-by-layer optimization. Notably, previously known methods like INTERP and FOURIER yield parameters that can be well fitted by linear functions, which supports our linearization strategy. Our analysis reveals that for the random Ising model, cost landscapes in this reduced parameter space demonstrate consistent structural patterns across different problem instances. Our experiments extend from classical simulation to actual quantum hardware implementation on IBM's Eagle processor, demonstrating the approach's viability on current NISQ devices. Furthermore, the numerical results indicate that parameter transferability primarily depends on the energy scale of problem instances, with normalization techniques improving transfer quality. Most of our numerical experiments are conducted on the random Ising model, while problem-dependence is also investigated across other models. A key advantage of parameter transferring is the complete elimination of instance-specific classical optimization overhead, as pre-trained parameters can be directly applied to other problem instances, reducing classical optimization costs by orders of magnitude for deeper circuits.

quant-ph↗

arXiv2025

Learning to Learn with Quantum Optimization via Quantum Neural Networks

Quantum Approximate Optimization Algorithms (QAOA) promise efficient solutions to classically intractable combinatorial optimization problems by harnessing shallow-depth quantum circuits. Yet, their performance and scalability often hinge on effective parameter optimization, which remains nontrivial due to rugged energy landscapes and hardware noise. In this work, we introduce a quantum meta-learning framework that combines quantum neural networks, specifically Quantum Long Short-Term Memory (QLSTM) architectures, with QAOA. By training the QLSTM optimizer on smaller graph instances, our approach rapidly generalizes to larger, more complex problems, substantially reducing the number of iterations required for convergence. Through comprehensive benchmarks on Max-Cut and Sherrington-Kirkpatrick model instances, we demonstrate that QLSTM-based optimizers converge faster and achieve higher approximation ratios compared to classical baselines, thereby offering a robust pathway toward scalable quantum optimization in the NISQ era.

quant-ph↗

arXiv2025

Enhancing NDAR with Delay-Gate-Induced Amplitude Damping

The Noise-Directed Adaptive Remapping (NDAR) method utilizes amplitude damping noise to enhance the performance of quantum optimization algorithms. NDAR alternates between exploration by sampling solutions from the quantum circuit and exploitation by transforming the cost Hamiltonian by changing the signs of its terms. Both exploration and exploitation are important components in classical heuristic algorithm design. In this study, we examine how NDAR performance improves by adjusting the balance between these components. We control the degree of exploitation by varying the delay time to 0, 50, and $100~μ\text{s}$, and investigate exploration strategies using two quantum circuits, QAOA and a random circuit, on IBM's Heron processor. Our results show that increasing delay time in NDAR improves the best objective value found in each iteration. In single-layer QAOA and random circuits applied to unweighted Max-Cut problem with low edge density, both exploration strategies yield similar objective value trajectories and provide competitive solution quality to simulated annealing for the 80-node problem. Their similar performance indicates that, in most cases, increasing amplitude damping noise via additional delay time results in information loss. On the other hand, QAOA outperforms random circuits in specific cases, such as positive-negative weighted Max-Cut on a fully connected graph. This suggests potential advantages of QAOA in more complex settings. We further develop a classical NDAR to better understand exploration strategies, demonstrating that controlling the Hamming weight distribution of sampled bitstrings yields higher quality solutions. This suggests that identifying suitable quantum circuits for exploration could enhance NDAR performance.

quant-ph↗

arXiv2025

Sampling-based Quantum Optimization Algorithm with Quantum Relaxation

Variational Quantum Algorithm (VQA) is a hybrid algorithm for noisy quantum devices. However, statistical fluctuations and physical noise degrade the solution quality, so it is difficult to maintain applicability for large-scale problems. In contrast, Sampling-based Quantum Algorithms have recently been successfully applied to large-scale quantum chemistry problems. The quantum device is used only for sampling, and the ground state and its energy are estimated on the classical device. In this study, we propose the Sampling-based Quantum Optimization Algorithm (SQOA). Two challenges exist in constructing a Sampling-based Quantum Algorithm for combinatorial optimization. The first challenge is that we need to encode the optimization problem in a non-diagonal Hamiltonian, even though many VQAs encode it into the Ising Hamiltonian, which is diagonal. The second challenge is that we need a method to prepare the input state to be sampled efficiently. We employ the Quantum Relaxation (QR) method for the first challenge, which encodes multiple classical variables in one qubit. It reduces required qubits compared to the Ising Hamiltonian approach. Moreover, we investigate the parameter transferability in the Quantum Alternating Operator Ansatz for QR Hamiltonians for the second challenge. We show that restricting parameters to a linear form exhibits moderate transferability for 3-regular MaxCut problems, similar to transferability observed in the Quantum Approximate Optimization Algorithm. This property allows us to efficiently prepare the input state for a large instance using the parameters from a small instance. We leveraged transferability to create input states and applied SQOA with QR to the MaxCut instances. Transferring parameters from a 20-node problem demonstrates that SQOA with QR provides high-quality solutions for 40-node problems without variational parameter optimization.

quant-ph↗

arXiv2025

Solving Capacitated Vehicle Routing Problem with Quantum Alternating Operator Ansatz and Column Generation

This study proposes a hybrid quantum-classical approach to solving the Capacitated Vehicle Routing Problem (CVRP) by integrating the Column Generation (CG) method with the Quantum Alternating Operator Ansatz (QAOAnsatz). The CG method divides the CVRP into the reduced master problem, which finds the best combination of the routes under the route set, and one or more subproblems, which generate the routes that would be beneficial to add to the route set. This method is iteratively refined by adding new routes identified via subproblems and continues until no improving route can be found. We leverage the QAOAnsatz to solve the subproblems. Our algorithm restricts the search space by designing the QAOAnsatz mixer Hamiltonian to enforce one-hot constraints. Moreover, to handle capacity constraints in QAOAnsatz, we employ an Augmented Lagrangian-inspired method that obviates the need for additional slack variables, reducing the required number of qubits. Experimental results on small-scale CVRP instances (up to 6 customers) show that QAOAnsatz converges more quickly to optimal routes than the standard QAOA approach, demonstrating the potential of this hybrid framework in tackling real-world logistical optimization problems on near-term quantum hardware.

quant-ph↗

arXiv2024

Quantum Local Search for Traveling Salesman Problem with Path-Slicing Strategy

We present novel path-slicing strategies integrated with quantum local search to optimize solutions for the Traveling Salesman Problem (TSP), addressing the limitations of current Noisy Intermediate-Scale Quantum (NISQ) technologies. Our hybrid quantum-classical approach leverages classical path initialization and quantum optimization to effectively manage the computational challenges posed by the TSP. We explore various path slicing methods, including k-means and anti-k-means clustering, to divide the TSP into manageable subproblems. These are then solved using quantum or classical solvers. Our analysis, performed on multiple TSP instances from the TSPlib, demonstrates the ability of our strategies to achieve near-optimal solutions efficiently, highlighting significant improvements in solving efficiency and resource utilization. This approach paves the way for future applications in larger combinatorial optimization scenarios, advancing the field of quantum optimization.

quant-ph↗

arXiv2024

Efficient Internal Strategies in Quantum Relaxation based Branch-and-Bound

A combinatorial optimization problem is to find an optimal solution under the constraints. This is one of the potential applications for quantum computers. Quantum Random Access Optimization (QRAO) is the quantum optimization algorithm that encodes multiple classical variables into a single qubit to construct a quantum Hamiltonian, thereby reducing the number of qubits required. The ground energy of the QRAO Hamiltonian provides a lower bound on the original problem's optimal value before encoding. This property allows the QRAO Hamiltonian to be used as a relaxation of the original problem, and it is thus referred to as a quantum relaxed Hamiltonian. In the Branch-and-Bound method, solving the relaxation problem plays a significant role. In this study, we developed Quantum Relaxation based Branch-and-Bound (QR-BnB), a method incorporating quantum relaxation into the Branch-and-Bound framework. We solved the MaxCut Problem and the Travelling Salesman Problem in our experiments. In all instances in this study, we obtained the optimal solution whenever we successfully computed the exact lower bound through quantum relaxation. Internal strategies, such as relaxation methods and variable selection, influence the convergence of the Branch-and-Bound. Thus, we have further developed the internal strategies for QR-BnB and examined how these strategies influence its convergence. We show that our variable selection strategy via the expectation value of the Pauli operators gives better convergence than the naive random choice. QRAO deals with only unconstrained optimization problems, but QR-BnB can handle constraints more flexibly because of the Branch-and-Bound processes on the classical computing part. We demonstrate that in our experiments with the Travelling Salesman Problem, the convergence of QR-BnB became more than three times faster by using the information in the constraints.

quant-ph↗

arXiv2024

Linearly simplified QAOA parameters and transferability

Quantum Approximate Optimization Algorithm (QAOA) provides a way to solve combinatorial optimization problems using quantum computers. QAOA circuits consist of time evolution operators by the cost Hamiltonian and of state mixing operators, and embedded variational parameter for each operator is tuned so that the expectation value of the cost function is minimized. The optimization of the variational parameters is taken place on classical devices while the cost function is measured in the sense of quantum. To facilitate the classical optimization, there are several previous works on making decision strategies for optimal/initial parameters and on extracting similarities among instances. In our current work, we consider simplified QAOA parameters that take linear forms along with the depth in the circuit. Such a simplification, which would be suggested from an analogy to quantum annealing, leads to a drastic reduction of the parameter space from 2p to 4 dimensions with the any number of QAOA layers p. In addition, cost landscapes in the reduced parameter space have some stability on differing instances. This fact suggests that an optimal parameter set for a given instance can be transferred to other instances. In this paper we present some numerical results that are obtained for instances of the random Ising model and of the max-cut problem. The transferability of linearized parameters is demonstrated for randomly generated source and destination instances, and its dependence on features of the instances are investigated.

quant-ph↗

arXiv2023

Anomalous fluctuations in homogeneous fluid phase of active Brownian particles

Giant number fluctuations (GNF) are an anomaly universally observed in active fluids with polar or nematic order. In this paper, we show that GNF arise in the fluid phase of active Brownian particles (ABP), where the polar order is absent. GNF in ABP extends over a large but finite length which characterizes the growing velocity correlations. To suppress unwanted phase separation and allow ones to explore the disordered fluid phase at large activities, we impart the inertia, or the mass, to the ABP. A linearized hydrodynamic theory captures our findings, but only qualitatively. We find numerically a nontrivial scaling relation for the density correlation function, which the linearized theory cannot explain. The results suggest ubiquitousness of the anomalous fluctuations even in the disordered homogeneous fluid phase in the absence of the directional order.

cond-mat.soft↗

arXiv2021

Anomalous transport phenomenon of a charged Brownian particle under the thermal gradient and the magnetic field

There is a growing interest in the stochastic processes of nonequilibrium systems subject to non-conserved forces, such as the magnetic forces acting on charged particles and the chiral self-propelled force acting on active particles. In this paper, we consider the stationary transport of non-interacting Brownian particles under a constant magnetic field in a position-dependent temperature background. We demonstrate the existence of the Nernst-like stationary density current perpendicular to both the temperature gradient and magnetic field, induced by the intricate coupling between the non-conserved force and the multiplicative noises due to the position-dependent temperature.

cond-mat.stat-mech↗

arXiv2021

Geometrical properties of mechanically annealed systems near the jamming transition

Geometrical properties of two-dimensional mixtures near the jamming transition point are numerically investigated using harmonic particles under mechanical training. The configurations generated by the quasi-static compression and oscillatory shear deformations exhibit anomalous suppression of the density fluctuations, known as hyperuniformity, below and above the jamming transition. For the jammed system trained by compression above the transition point, the hyperuniformity exponent increases. For the system below the transition point under oscillatory shear, the hyperuniformity exponent also increases until the shear amplitude reaches the threshold value. The threshold value matches with the transition point from the point-reversible phase where the particles experience no collision to the loop-reversible phase where the particles' displacements are non-affine during a shear-cycle before coming back to an original position. The results demonstrated in this paper are explained in terms of neither of universal criticality of the jamming transition nor the nonequilibrium phase transitions.

cond-mat.soft↗

arXiv2018

The Entropy Anomaly and the Linear Irreversible Thermodynamics

The irreversible currents and entropy production rate of a dilute colloidal suspension are calculated using the linear irreversible thermodynamics and the linear response theory. The \anomalous" or \hidden" entropy recently discussed in the context of the stochastic thermodynamics is fully accounted in these classic frameworks. We show that the two distinct formulations lead to identical results as long as the local equilibrium assumption or, equivalently the linear response theory, is valid.

cond-mat.stat-mech↗