betaFeedback

WorksPhysics and Astronomy

A Variational Quantum Algorithm for Nonlinear Finite Element Analysis of Hyperelastic Materials

Abstract

This manuscript explores a variational quantum formulation for nonlinear elasticity problems arising from hyperelastic material models. The approach leverages the potential energy structure of hyperelasticity and employs a hybrid quantum classical framework in which the energy functional is evaluated using parameterized quantum circuits and optimized through classical routines. To enable a hybrid (classical quantum implementation), polynomial approximations of the nonlinear terms in strain energy density are introduced, yielding a representation compatible with variational quantum algorithms. The methodology is demonstrated on a special case of the NeoHookean material model in a one dimensional setting using finite element discretizations with first and second order shape functions and nonhomogeneous boundary conditions. Numerical experiments investigate the influence of the polynomial approximation order on the accuracy and efficiency of the proposed approach, illustrating its feasibility for near-term quantum devices.

Introduction

Numerical evaluation of nonlinear partial differential equations (PDEs) sits at the foundation of computational mechanics. A common approach to solving nonlinear mechanics problems is to obtain an algebraic system following discretization, which is then solved iteratively. The computational cost of the overall process grows rapidly with the system size. As demand for simulation of large, complex material and structural systems increases, such computations place significant pressure on conventional computing architectures, particularly due to performance, memory, and energy-efficiency limitations associated with the von Neumann computing architectures [1]. Quantum computing (QC) has recently emerged as an alternative computing paradigm, owing to the exponential growth of its accessible state space with the number of qubits [2, 3, 4]. In this work, we take an initial step in developing and investigating a variational quantum algorithm directed toward problems in nonlinear elasticity by considering a special case of the Neo-Hookean material model as a representative case.

The majority of the relevant existing QC literature focuses on the evaluation of linear ordinary differential equations (ODEs) and PDEs. A common solution approach is to construct a linear system of equations following discretization, and evaluating it using quantum linear systems algorithms (QLSAs) [5, 6]. The foundational work by Harrow et al. [7] (HHL) introduced the first QLSA, which solves sparse linear systems Ax=bA x = b by encoding the right-hand side vector bb as a quantum state ∣b⟩\lvert b\rangle. Using Hamiltonian simulation [8] and quantum phase estimation, the algorithm prepares a state ∣x⟩∝A−1∣b⟩\lvert x\rangle\propto A^{-1}\lvert b\rangle that encodes the properties of the solution vector xx. Subsequent developments produced high-precision QLSA variants based on linear combination of unitaries (LCU) [9], quantum singular value transformation (QSVT) [10], and adiabatic approaches [11]. However, despite their favorable asymptotic complexity, the aforementioned approaches require deep circuits and error rates that are beyond the capabilities of the NISQ-era hardware [12].

Within the framework of near-term quantum computing, Variational Quantum Algorithms (VQAs) have emerged as a prominent and widely explored class of algorithms [13, 14, 15]. The central idea behind VQAs is to offload the computationally intensive parts of a problem to a quantum processor that evaluates a cost function using parameterized quantum circuits encoding the quantity of interest. The circuit parameters are then iteratively updated by a classical optimizer to minimize this cost function. VQAs face well-documented trainability challenges. As circuit width or ansatz depth increases, the cost landscape can develop barren plateaus, regions where gradients vanish exponentially and classical optimization stalls [16]; this behavior is closely tied to the choice of cost function and the ansatz, with global observables inducing barren plateaus more readily than local ones [17]. The barren plateau problem is compounded by hardware noise [18].

Several efforts have applied general-purpose quantum algorithms to problems in computational mechanics. For example, the Poisson equation and elliptic finite element problems have been addressed via QLSAs [19, 20], lying in the fault-tolerant quantum computing regime. Similarly, Liu et al. [21] demonstrate an algorithm for the linear elastic homogenization of periodic composites using a quantum Fourier transform (QFT)-based approach. Within the VQA regime, the Variational Quantum Linear Solver (VQLS) [22] emerged as one of the prominent VQAs for linear systems, and was used by Arora et al. [23] for FEM-based steady-state heat conduction, and Rao et al. [24] for plane strain elasticity problems. Raisuddin and De [25] instead employed quantum annealers to solve linear finite element systems via an equivalent quadratic optimization formulation.

The extension of quantum algorithms to nonlinear problems in mechanics remains challenging due to the inherently linear structure of quantum operations. Several existing approaches reformulate nonlinearities in the governing equations into representations compatible with available quantum algorithmic primitives, differing primarily in the quantities being expanded and the resulting reformulation. Kyriienko et al. [26] represent the solution function through a truncated Chebyshev quantum feature map, while the nonlinear operations associated with the governing equation are evaluated through classical post-processing. Xu et al. [27] expand the solution path with respect to a continuation parameter, resulting in a sequence of linear subproblems solved using VQLS. Endo and Takahashi [28] employ Carleman linearization to embed a nonlinear differential equation into a larger truncated linear system. Soize et al. [29] apply a similar strategy to a Fokker–Planck eigenvalue problem, which, although linear, involves an underlying non-algebraic potential approximated via a truncated polynomial expansion to enable a Variational Quantum Eigensolver-based solution.

In the VQA regime, Lubasch et al. [30, 31] introduced a novel VQA framework for nonlinear PDEs with polynomial nonlinearities by devising quantum nonlinear processing units and demonstrated the efficacy of this approach on Burgers and nonlinear Schrödinger equations. Sarma et al. [32] recently extended this approach to solve a broader class of nonlinear PDEs. The idea expressed in these works relies on a least-squares functional as the cost function. Sato et al. [33] proposed a VQA to solve the Poisson equation by formulating the corresponding Dirichlet energy as a cost function parameterized by a quantum circuit. The energy terms were evaluated using a quantum oracle, and the circuit parameters were optimized classically to minimize the cost, thereby encoding the solution. Similarly, Over and Bengoechea et al. [34, 35], also derived a variational form for the advection–diffusion equation in the time-discrete setting and minimized it within a VQA framework.

This manuscript presents a VQA-based approach for nonlinear elasticity problems arising from hyperelastic material models under certain assumptions. The proposed framework exploits the fact that, for hyperelastic materials, the stress is derived from a strain energy density function, thereby enabling a variational formulation in terms of a total potential energy functional [36, 37, 38, 39, 40]. The resulting approximated potential energy functional is constructed to integrate naturally within a variational quantum framework. The proposed methodology is demonstrated on a simplified one-dimensional Neo-Hookean material-based problem. Two distinct approximation strategies are developed and implemented in accordance with contemporary VQA capabilities. The displacement field is discretized using first- and second-order finite element shape functions, and the formulation accommodates nonhomogeneous boundary conditions. The influence of higher-order polynomial approximations on the efficiency and accuracy of the VQA implementation is evaluated. Numerical verification of the proposed approach has been performed using noise-free statevector simulations rather than execution on real quantum hardware. We discuss several practical algorithmic and hardware-based challenges towards a practical realization of the framework and extension to multidimensional problems. The novel contributions of the proposed approach are therefore as follows: (1) The numerical problem is expressed as the minimization of the variational form (as opposed to the weak form) to prepare it for VQA; and (2) treatment of the nonlinear terms in the variational form using the QNPUs.

The remainder of this manuscript is organized as follows. § 2 presents the overview of the proposed approach. The governing equations for a general multidimensional problem are first introduced and then restricted to the one-dimensional setting. The resulting potential energy is then embedded into a VQA framework, concluding with an outline of the overall VQA procedure. § 3 applies the scheme to a special case of the Neo-Hookean material model, introducing the energy functional, deriving a quantum-computing-compatible approximation, and constructing its discretized form using finite element shape functions and Gaussian quadrature. § 4 provides the details of the quantum implementation. The basic quantum circuits used as building blocks, the state preparation procedure for encoding real-valued vectors, and the evaluation of the individual terms in the cost function based on specifically designed quantum circuits are described. In § 5, we analyze the gate and time complexity of the proposed approach. § 7 presents numerical results demonstrating the performance of the VQA for different polynomial approximation orders and for first- and second-order finite element discretizations. § 8 discusses the practical challenges associated with realizing the proposed framework and outlines potential future directions for addressing some of these issues. § 9 summarizes the main findings and provides a conclusion.

Overview of the Proposed Approach

In this section, we outline the proposed VQA-based approach for nonlinear elasticity problems involving hyperelastic materials. We begin by formulating the governing equations and deriving a simplified model suitable for analysis. The VQA framework requires a cost function to optimize, which we construct from the system’s potential energy. To evaluate this energy on a quantum computer, the displacement field (represented by its discretized values) must be encoded as a quantum state. This motivates the introduction of a quantum representation of real-valued vectors. Finally, we integrate these components to establish a complete framework for approximating the solution to the underlying elasticity problem.

Mathematical model

The nonlinear elastostatic boundary-value problem in the reference configuration Ω0⊂R3\Omega_0 \subset\mathbb{R}^3 (X=(X,Y,Z)\mathbf{X} = (X,Y,Z)) is given by

∇X⋅P(F(X))+B(X)=0for X∈Ω0,(1)\nabla_{\mathbf{X}} \cdot\boldsymbol{P}(\boldsymbol{F}(\mathbf{X})) + \boldsymbol{B}(\mathbf{X}) = \mathbf{0} \quad\text{for } \mathbf{X} \in\Omega_0, \tag*{(1)}
F=∇Xφ=I+∇Xu(X)(2)\boldsymbol{F} = \nabla_{\mathbf{X}}\boldsymbol{\varphi} = \boldsymbol{I} + \nabla_{\mathbf{X}}\boldsymbol{u}(\mathbf{X}) \tag*{(2)}
u(X)=u‾(X)for X∈ΓD,(3)\boldsymbol{u}(\mathbf{X}) = \overline{\boldsymbol{u}}(\mathbf{X}) \quad\text{for } \mathbf{X} \in\Gamma_D, \tag*{(3)}
P⋅N=P‾(X)for X∈ΓN,(4)\boldsymbol{P} \cdot\boldsymbol{N} = \overline{\boldsymbol{P}}(\mathbf{X}) \quad\text{for } \mathbf{X} \in\Gamma_N, \tag*{(4)}

where P\boldsymbol{P} is the first Piola-Kirchhoff stress tensor, φ\boldsymbol{\varphi} the deformation map, u=(u,v,w)\boldsymbol{u} = (u,v,w) the displacement, F\boldsymbol{F} the deformation gradient and B\boldsymbol{B} the body force per unit reference volume. For a second-order tensor field T(X)\boldsymbol{T}(\mathbf{X}), (∇X⋅T)a=∂TaA/∂XA(\nabla_{\mathbf{X}} \cdot\boldsymbol{T})_a = \partial T_{aA}/\partial X_A [39, 37]. N\boldsymbol{N} is the outward unit normal at the reference domain boundaries, and u‾\overline{\boldsymbol{u}} and P‾\overline{\boldsymbol{P}} are prescribed displacements and tractions, respectively. The Dirichlet and Neumann boundaries are denoted as ΓD⊂∂Ω0\Gamma_D \subset\partial\Omega_0 and ΓN⊂∂Ω0\Gamma_N \subset\partial\Omega_0, respectively, such that ΓD∪ΓN=∂Ω0\Gamma_D \cup\Gamma_N = \partial\Omega_0 and ΓD∩ΓN=∅\Gamma_D \cap\Gamma_N = \varnothing. For hyperelastic materials, the stress follows from a strain energy density W(F)W(\boldsymbol{F}) via P=∂W/∂F\boldsymbol{P} = \partial W/\partial\boldsymbol{F}. For such materials, it is a classical result that there exists a potential energy functional

E[u]=∫Ω0W(F) dV−∫Ω0B⋅u dV−∫ΓNP‾⋅u dA,(5)\mathcal{E}[\boldsymbol{u}] = \int_{\Omega_0} W(\boldsymbol{F})\,\mathrm{d}V - \int_{\Omega_0} \boldsymbol{B}\cdot\boldsymbol{u}\,\mathrm{d}V - \int_{\Gamma_N} \overline{\boldsymbol{P}}\cdot\boldsymbol{u}\,\mathrm{d}A, \tag*{(5)}

up to a constant that depends on the applied traction and the body force. The critical points of E\mathcal{E} satisfy the weak form of the governing equations (1). The number and nature of the critical points (i.e., minima vs. saddle points) of the potential energy functional depend on the properties of WW [36].

We consider a one-dimensional reduction of the above three-dimensional theory. Specifically, we assume that the reference configuration is a prismatic body of constant cross-sectional area AA, such that

Ω0=(0,L)×A,(6)\Omega_0 = (0,L) \times A, \tag*{(6)}

where A⊂R2A \subset\mathbb{R}^2. We further assume that all fields depend only on the axial coordinate XX, and that the deformation is uniaxial along the XX-direction. The lateral surface (0,L)×∂A(0,L) \times\partial A is assumed to be traction-free, while the remainder of the boundary conditions are prescribed on the end faces at X=0X = 0 and X=LX = L.

Under these assumptions, the deformation gradient reduces to

F(X)=diag⁡(λ(X),1,1),(7)\boldsymbol{F}(X) = \operatorname{diag}(\lambda(X),1,1), \tag*{(7)}

where

λ(X)=1+u′(X)(8)\lambda(X) = 1 + u'(X) \tag*{(8)}

represents the principal stretch in the axial direction and (⋅)′≡d(⋅)/dX(\cdot)' \equiv\mathrm{d}(\cdot)/\mathrm{d}X. Physical admissibility requires

J(X)=det⁡F(X)=λ(X)>0,∀X∈(0,L)(9)J(X) = \det\boldsymbol{F}(X) = \lambda(X) > 0, \quad\forall X \in(0,L) \tag*{(9)}

Since all fields are uniform over the cross-section, volume and surface integrals factor into the cross-sectional area AA, multiplied by one-dimensional integrals over (0,L)(0,L). For simplicity, and without loss of generality, we work per unit reference cross-sectional area (i.e., ∣A∣=1|A|=1), so that the three-dimensional formulation reduces to a one-dimensional problem on the interval

Ω0=(0,L),∂Ω0={0,L}.(10)\Omega_0=(0,L), \qquad\partial\Omega_0=\{0,L\}. \tag*{(10)}

The Dirichlet and Neumann boundaries therefore correspond to subsets of the endpoints. Consequently, we consider the one-dimensional reduction of the potential energy functional: (2):

E[u]:=∫Ω0W(λ) dX−∫Ω0B(X)u(X) dX−∑X∈ΓNPˉu(X),(11)\mathcal{E}[u] := \int_{\Omega_0} W(\lambda)\,\mathrm{d}X-\int_{\Omega_0}B(X)u(X)\,\mathrm{d}X-\sum_{X\in\Gamma_N}\bar{P}u(X), \tag*{(11)}

where BB represents the body force, Pˉ\bar{P} the prescribed axial traction and WW is a function of λ\lambda only. The stationary point of the above functional yields the following weak form:

For u=uˉu=\bar{u} on ΓD\Gamma_D, find uu such that

∫Ω0P(λ)v′ dX=∫Ω0Bv dX+∑X∈ΓNPˉv,∀v∈H1(Ω0);v=0 on ΓD(12)\int_{\Omega_0}P(\lambda)v'\,\mathrm{d}X=\int_{\Omega_0}Bv\,\mathrm{d}X+\sum_{X\in\Gamma_N}\bar{P}v,\qquad\forall v\in H^1(\Omega_0);\quad v=0\text{ on }\Gamma_D \tag*{(12)}

where P(λ)=∂W/∂λP(\lambda)=\partial W/\partial\lambda.

Under sufficient smoothness assumptions on the displacement field, the weak form is equivalent to the strong form (equilibrium equation) in the axial direction, given by:

ddX(P(λ(X)))+B(X)=0in Ω0,(13)\frac{\mathrm{d}}{\mathrm{d}X}\bigl(P(\lambda(X))\bigr)+B(X)=0\quad\text{in }\Omega_0, \tag*{(13)}

subject to the boundary conditions

u=uˉon ΓD,P=Pˉon ΓN.(14)u=\bar{u}\quad\text{on }\Gamma_D,\qquad P=\bar{P}\quad\text{on }\Gamma_N. \tag*{(14)}

Quantum representation of real vectors

Unlike classical computing where information is represented using bits that take values of 0 or 1, quantum computing uses qubits—quantum two-level systems—that can exist in a superposition of the 0 and 1 states. The computational basis states for a single qubit are denoted as ∣0⟩|0\rangle and ∣1⟩|1\rangle. For a register of nn qubits, the corresponding Hilbert space has dimension Nq=2nN_q=2^n. The computational basis states are denoted as

{∣bin(k)⟩∣k=0,1,…,2n−1},(15)\{|{\rm bin}(k)\rangle\mid k=0,1,\ldots,2^n-1\}, \tag*{(15)}

where kk is the integer label of the basis state, and its binary representation specifies the state of each qubit. For example, in a three-qubit system, ∣bin(5)⟩|{\rm bin}(5)\rangle corresponds to the quantum state ∣101⟩|101\rangle, which indicates that the first and third qubits are in state ∣1⟩|1\rangle, while the second qubit is in state ∣0⟩|0\rangle. In what follows, we use ∣k⟩|k\rangle instead of ∣bin(k)⟩|{\rm bin}(k)\rangle for simplicity, unless otherwise stated.

Any real vector p∈RNq\mathbf{p}\in\mathbb{R}^{N_q} (complex vectors are admissible, but we restrict our attention to real-valued vectors), is encoded as a quantum state ∣p^⟩|\hat{p}\rangle (or simply p^\hat{p} when no confusion arises), such that:

∑i=0Nq−1piei=p=∥p∥ ∣p^⟩=∥p∥∑k=0Nq−1p^k∣k⟩,(16)\sum_{i=0}^{N_q-1}p_i\mathbf{e}_i=\mathbf{p}=\|\mathbf{p}\|\,|\hat{p}\rangle=\|\mathbf{p}\|\sum_{k=0}^{N_q-1}\hat{p}_k|k\rangle, \tag*{(16)}

noting that the quantum state ∣p^⟩|\hat{p}\rangle is normalized.

Any quantum state ∣p^⟩|\hat{p}\rangle is prepared on a quantum computer using a parameterized quantum circuit, or Ansatz circuit, represented by a unitary operator V^(λp)\hat{V}(\lambda_p), where

λp=(λp1,λp2,…,λpm);λpi∈[0,2π],(17)\lambda_p = (\lambda_{p1},\lambda_{p2},\ldots,\lambda_{pm}); \quad\lambda_{pi} \in[0,2\pi], \tag*{(17)}

is the set of ansatz parameters (each element corresponds to a rotation angle applied to a qubit). When applied to the initial basis state ∣0⟩|0\rangle, it yields:

∣p^⟩=V^(λp)∣0⟩.(18)|\hat{p}\rangle= \hat{V}(\lambda_p)|0\rangle. \tag*{(18)}

The ansatz defines a parameterized trial space of quantum states, whose structure determines its ability to approximate the target state. Different ansatz constructions employ specific arrangements of single-qubit rotation gates and two-qubit entangling gates, which govern the expressiveness of the resulting state space. In this work, we adopt the ansatz proposed by Sim et al. [41], shown in Figure 1. Other ansatz choices with sufficient expressiveness could be used in its place without loss of generality. Ansatz parameters along with the scaling parameter λp0:=∥p∥\lambda_{p0} := \|\boldsymbol{p}\|, are collectively referred to as the control parameters and generate the vector p\boldsymbol{p} based on (13).

VQA Framework

Figure 1. VQA Framework.

Variational Quantum Algorithm

The overall structure of the proposed VQA framework is depicted in Figure 1. The workflow consists of two main phases: a setup phase (purple) and an optimization phase (blue). At its core, the method represents the unknown displacement field through a set of tunable control parameters, which are iteratively adjusted to approximate the solution.

In the setup phase, the elasticity problem is specified via its strain energy density, which defines the corresponding potential energy functional. This functional is considered for evaluation using quantum circuits, with the control parameters serving as circuit input. Instead of assembling element-wise contributions, the quantum approach directly estimates global or macroscopic quantities (expectation values) of interest. In particular, expectation values from a small set of circuits, are combined classically to estimate the total potential energy of the system.

A key challenge, however, arises from the inherently linear nature of quantum operations, which makes the direct implementation of nonlinear transformations involved in the potential energy difficult. Existing methods for nonlinear transformations can be resource-intensive [42]. Instead, we propose low-degree polynomial approximations of arbitrary nonlinearities, which can be implemented using relatively shallow quantum circuits known as Quantum Nonlinear Processing Units (QNPUs), as developed in [30], and are more suitable for near-term NISQ devices.

Accordingly, any nonlinearity G(u′)\mathcal{G}(u') in the functional (11) is approximated as

G(u′)≈∑i=0Ntnci(u′)i,(19)\mathcal{G}(u') \approx\sum_{i=0}^{N_{\mathrm{tn}}} c_i (u')^i, \tag*{(19)}

References

  1. [1]Kaijie Wei et al. “A data compressor for FPGA-based state vector quantum simulators”. In: Proceedings of the 14th International Symposium on Highly Efficient Accelerators and Reconfigurable Technologies. HEART ’24. Porto, Portugal: Association for Computing Machinery, 2024, pp. 63–70. ISBN: 978400717277. DOI: 10.1145/3665283.3665293. URL: https://doi.org/10.1145/3665283.3665293.
  2. [2]Zebo Yang, Maede Zolanvari, and Raj Jain. “A Survey of Important Issues in Quantum Computing and Communications”. In: IEEE Communications Surveys & Tutorials 25.2 (2023), pp. 1059–1100. DOI: 10.1109/COMST.2023.3241234.DOI
  3. [3]Andrew M. Childs, Jin-Peng Liu, and Aaron Ostrander. “Improved Quantum Algorithms for Linear and Nonlinear Differential Equations”. In: Quantum 7 (2023), p. 913. DOI: 10.22331/q-2023-02-02-913.
  4. [4]Giorgio Tosti Balducci et al. “Review and perspectives in quantum computing for partial differential equations in structural mechanics”. In: Frontiers in Mechanical Engineering Volume 8 - 2022 (2022). ISSN: 2297-3079. DOI: 10.3389/fmech.2022.914241. URL: https://www.frontiersin.org/journals/mechanical-engineering/articles/10.3389/fmech.2022.914241.DOI
  5. [5]Dominic W. Berry. Quantum algorithm for linear differential equations. arXiv:1701.03684. 2017.
  6. [6]Jin-Peng Liu. “Quantum Algorithms for Linear and Nonlinear Differential Equations”. PhD thesis. University of Maryland, 2022.DOI
  7. [7]Aram W. Harrow, Avinatan Hassidim, and Seth Lloyd. “Quantum Algorithm for Linear Systems of Equations”. In: Physical Review Letters 103.15 (2009), p. 150502. DOI: 10.1103/PhysRevLett.103.150502.DOI
  8. [8]Seth Lloyd. “Universal Quantum Simulators”. In: Science 273.5278 (1996), pp. 1073–1078. DOI: 10.1126/science.273.5278.1073.DOI
  9. [9]Dominic W. Berry and Andrew M. Childs. “Hamiltonian simulation using linear combinations of unitary operations”. In: Quantum Information and Computation 12 (2012), pp. 29–62.
  10. [10]John M. Martyn et al. “A Grand Unification of Quantum Algorithms”. In: PRX Quantum 2 (2021), p. 040203. DOI: 10.1103/PRXQuantum.2.040203.DOI
  11. [11]Yigit Subasi, Rolando D. Somma, and Davide Orsucci. “Quantum Algorithms for Systems of Linear Equations Inspired by Adiabatic Quantum Computing”. In: Physical Review Letters 122.6 (2019), p. 060504. DOI: 10.1103/PhysRevLett.122.060504.DOI
  12. [12]John Preskill. “Quantum Computing in the NISQ era and beyond”. In: Quantum 2 (2018), p. 79. DOI: 10.22331/q-2018-08-06-79.DOI
  13. [13]Kishor Bharti et al. “Noisy intermediate-scale quantum algorithms”. In: Rev. Mod. Phys. 94 (1 Feb. 2022), p. 015004. DOI: 10.1103/RevModPhys.94.015004. URL: https://link.aps.org/doi/10.1103/RevModPhys.94.015004.DOI
  14. [14]Alberto Peruzzo et al. “A variational eigenvalue solver on a photonic quantum processor”. In: Nature Communications 5.1 (2014), p. 4213. DOI: 10.1038/ncomms5213. URL: https://doi.org/10.1038/ncomms5213.
  15. [15]M. Cerezo et al. “Variational Quantum Algorithms”. In: Nature Reviews Physics 3 (2021), pp. 625–644. DOI: 10.1038/s42254-021-00348-9.DOI
  16. [16]Jarrod R. McClean et al. “Barren plateaus in quantum neural network training landscapes”. In: Nature Communications 9 (2018), p. 4812. DOI: 10.1038/s41467-018-07090-4.DOI
  17. [17]M. Cerezo et al. “Cost function dependent barren plateaus in shallow parametrized quantum circuits”. In: Nature Communications 12 (2021), p. 1791. DOI: 10.1038/s41467-021-21728-w.DOI
  18. [18]Samson Wang et al. “Noise-induced barren plateaus in variational quantum algorithms”. In: Nature Communications 12 (2021), p. 6961. DOI: 10.1038/s41467-021-27045-6.DOI
  19. [19]Yudong Cao et al. “Quantum algorithm and circuit design solving the Poisson equation”. In: New Journal of Physics 15.1 (2013), p. 013021. DOI: 10.1088/1367-2630/15/1/013021.DOI
  20. [20]Ashley Montanaro and Sam Pallister. “Quantum algorithms and the finite element method”. In: Physical Review A 93.3 (2016), p. 032324. DOI: 10.1103/PhysRevA.93.032324.DOI
  21. [21]Burigede Liu, Michael Ortiz, and Fehmi Cirak. “Towards quantum computational mechanics”. In: Computer Methods in Applied Mechanics and Engineering 432 (2024), p. 117403. ISSN: 0045-7825. DOI: https://doi.org/10.1016/j.cma.2024.117403. URL: https://www.sciencedirect.com/science/article/pii/S0045782524006583.
  22. [22]Carlos Bravo-Prieto et al. “Variational Quantum Linear Solver”. In: Quantum 7 (Nov. 2023), p. 1188. DOI: 10.22331/q-2023-11-22-1188. URL: https://doi.org/10.22331/q-2023-11-22-1188.
  23. [23]A. Arora, B. M. Ward, and C. Oskay. “An implementation of the finite element method in hybrid classical/quantum computers”. In: Finite Elements in Analysis and Design 248 (2025), p. 104354.arxiv.org/abs/2411.09038
  24. [24]Xiang Rao and Kou Du. “Performance Study of Variational Quantum Linear Solver for Linear Elastic Problems”. In: Computational and Experimental Simulations in Engineering. Ed. by Kun Zhou. Vol. 168. Mechanisms and Machine Science. Cham: Springer, 2024, pp. 80–94. ISBN: 978-3-031-68775-4. DOI: 10.1007/978-3-031-68775-4_6.DOI
  25. [25]Osama Muhammad Raisuddin and Suvranu De. “FEQa: Finite element computations on quantum annealers”. In: Computer Methods in Applied Mechanics and Engineering 395 (2022), p. 115014. ISSN: 0045-7825. DOI: https://doi.org/10.1016/j.cma.2022.115014. URL: https://www.sciencedirect.com/science/article/pii/S0045782522002444.
  26. [26]Oleksandr Kyriienko, Annie E. Paine, and Vincent E. Elfving. “Solving nonlinear differential equations with differentiable quantum circuits”. In: Phys. Rev. A 103 (5 May 2021), p. 052416. DOI: 10.1103/PhysRevA.103.052416. URL: https://link.aps.org/doi/10.1103/PhysRevA.103.052416.DOI
  27. [27]Yongchun Xu et al. “A robust quantum nonlinear solver based on the asymptotic numerical method”. In: arXiv preprint arXiv:2412.03939 (2024). DOI: 10.48550/arXiv.2412.03939. URL: https://arxiv.org/abs/2412.03939.
  28. [28]Katsuhiro Endo and Kazuaki Z. Takahashi. “Divergence-free algorithms for solving nonlinear differential equations on quantum computers”. In: Phys. Rev. Res. 8 (1 Jan. 2026), p. 013057. DOI: 10.1103/2hfp-qyz1. URL: https://link.aps.org/doi/10.1103/2hfp-qyz1.
  29. [29]Christian Soize, Loïc Joubert-Doriol, and Artur F. Izmaylov. “Quantum computer formulation of the FKP-operator eigenvalue problem for probabilistic learning on manifolds”. In: Computer Methods in Applied Mechanics and Engineering 443 (2025), p. 118080. ISSN: 0045-7825. DOI: https://doi.org/10.1016/j.cma.2025.118080. URL: https://www.sciencedirect.com/science/article/pii/S0045782525003524.
  30. [44]Milton Abramowitz and Irene A. Stegun. Handbook of Mathematical Functions with Formulas, Graphs, and Mathematical Tables. Vol. 55. Applied Mathematics Series. Washington, D.C.: National Bureau of Standards, 1964.
  31. [45]Karel Rektorys. Variational Methods in Mathematics, Science and Engineering. 2nd ed. Dordrecht: D. Reidel Publishing Company, 1977.DOI
  32. [46]Michael A. Nielsen and Isaac L. Chuang. Quantum Computation and Quantum Information: 10th Anniversary Edition. Cambridge, United Kingdom: Cambridge University Press, 2010. ISBN: 978-1-107-00217-3.
  33. [47]Robert Beals et al. “Quantum lower bounds by polynomials”. In: J. ACM 48.4 (July 2001), pp. 778–797. ISSN: 0004-5411. DOI: 10.1145/502090.502097. URL: https://doi.org/10.1145/502090.502097.
  34. [48]Daan Camps et al. “Explicit Quantum Circuits for Block Encodings of Certain Sparse Matrices”. In: SIAM Journal on Matrix Analysis and Applications 45.1 (Mar. 2024). DOI: 10.1137/22m1484298.DOI
  35. [49]Tillman Philo and Caglar Oskay. “PHASE: Pauli Hierarchical Assembly on Subdivided Elements for Quantum-Compatible Operator Synthesis”. In: arXiv preprint arXiv:2606.11478 (2026).arxiv.org/abs/2606.11478
  36. [50]Ali Javadi-Abhari et al. “Quantum computing with Qiskit”. In: arXiv preprint arXiv:2405.08810 (2024).
  37. [51]Maria Schuld et al. “Evaluating analytic gradients on quantum hardware”. In: Phys. Rev. A 99 (3 Mar. 2019), p. 032331. DOI: 10.1103/PhysRevA.99.032331. URL: https://link.aps.org/doi/10.1103/PhysRevA.99.032331.DOI
  38. [52]R. Fletcher. “A new approach to variable metric algorithms”. In: The Computer Journal 13.3 (1970), pp. 317–322.DOI
  39. [53]J. Nocedal and S. J. Wright. Numerical Optimization. 2nd ed. Springer, 2006.
  40. [54]Mikko Möttönen et al. “Transformation of Quantum States Using Uniformly Controlled Rotations”. In: Quantum Information & Computation 5.6 (2005), pp. 467–473. DOI: 10.26421/QIC5.6-5.DOI
  41. [55]Debora Ramacciotti, Andrea I. Lefterovici, and Antonio F. Rotundo. “Simple quantum algorithm to efficiently prepare sparse states”. In: Phys. Rev. A 110 (3 Sept. 2024), p. 032609. DOI: 10.1103/PhysRevA.110.032609. URL: https://link.aps.org/doi/10.1103/PhysRevA.110.032609.DOI
  42. [56]Oliver O’Brien and Christoph Sünderhauf. “Quantum state preparation via piecewise QSVT”. In: Quantum 9 (July 2025), p. 1786. ISSN: 2521-327X. DOI: 10.22331/q-2025-07-03-1786. URL: https://doi.org/10.22331/q-2025-07-03-1786.
  43. [57]Alok Shukla and Prakash Vedula. “An efficient quantum algorithm for preparation of uniform quantum superposition states”. In: Quantum Information Processing 23.1 (2024), p. 38. DOI: 10.1007/s11128-024-04258-4. arXiv: 2306.11747.DOI
  44. [58]Gilles Brassard et al. “Quantum Amplitude Amplification and Estimation”. In: Quantum Computation and Information. Ed. by Samuel J. Lomonaco and Howard E. Brandt. Vol. 305. Contemporary Mathematics. Providence, RI: American Mathematical Society, 2002, pp. 53–74.DOI
  45. [59]Stephen Abbott. Understanding Analysis. 2nd. Undergraduate Texts in Mathematics. New York: Springer, 2015. ISBN: 978-1-4939-2711-1. DOI: 10.1007/978-1-4939-2712-8.DOI

Paper details

vlnc:W_9hh27p8e
Version
v1 · latest
Published
Fresh — added to the Hub within the last 30 days
Added
License
CC BY 4.0
Usage

views · 0

downloads · 0

A view counts once per browser per day, a download each time the PDF is requested.

  • Physics and Astronomy

Not recorded for this paper

View statement
Looking for related papers…
Contents