Abstract
This paper is devoted to the derivation of a digital quantum algorithm for the Cauchy problem for symmetric first-order linear hyperbolic systems, thanks to the reservoir technique. The reservoir technique is a method designed to avoid artificial diffusion generated by first-order finite volume methods approximating hyperbolic systems of conservation laws. For some class of hyperbolic systems, namely, those with constant matrices in several dimensions, we show that the combination of (i) the reservoir method and (ii) the alternate direction iteration operator splitting approximation allows for the derivation of algorithms only based on simple unitary transformations, thus being perfectly suitable for an implementation on a quantum computer. The same approach can also be adapted to scalar one-dimensional systems with non-constant velocity by combining with a non-uniform mesh. The asymptotic computational complexity for the time evolution is determined and it is demonstrated that the quantum algorithm is more efficient than the classical version. However, in the quantum case, the solution is encoded in probability amplitudes of the quantum register. As a consequence, as with other similar quantum algorithms, a post-processing mechanism has to be used to obtain general properties of the solution because a direct reading cannot be performed as efficiently as the time evolution.
Similar content being viewed by others
References
Abrams, D.S., Lloyd, S.: Quantum algorithm providing exponential speed increase for finding eigenvalues and eigenvectors. Phys. Rev. Lett. 83, 5162–5165 (1999)
Aharonov, D., Ta-Shma, A.: Adiabatic quantum state generation and statistical zero knowledge. In: Proceedings of the thirty-fifth annual ACM symposium on theory of computing, pp. 20–29, ACM (2003)
Alouges, F., De Vuyst, F., Le Coq, G., Lorin, E.: A process of reduction of the numerical diffusion of usual order one flux difference schemes for nonlinear hyperbolic systems [un procédé de réduction de la diffusion numérique des schémas à différence de flux d’ordre un pour les systèmes hyperboliques non linéaires]. C.R. Math. 335(7), 627–632 (2002)
Alouges, F., De Vuyst, F., Le Coq, G., Lorin, E.: The reservoir scheme for systems of conservation laws. In: Finite volumes for complex applications, III (Porquerolles, 2002), pp. 247–254. Hermes Sci. Publ., Paris (2002)
Alouges, F., De Vuyst, F., Le Coq, G., Lorin, E.: The reservoir technique: a way to make Godunov-type schemes zero or very low diffuse. application to Colella-Glaz solver. Eur. J. Mech. B. Fluids 27(6), 643–664 (2008)
Alouges, F., Le Coq, G., Lorin, E.: Two-dimensional extension of the reservoir technique for some linear advection systems. J. of Sc. Comput. 31(3), 419–458 (2007)
Arrighi, P., Nesme, V., Forets, M.: The Dirac equation as a quantum walk: higher dimensions, observational convergence. J. Phys. A Math. Theor. 47(46), 465302 (2014)
Aspuru-Guzik, A., Dutoi, A.D., Love, P.J., Head-Gordon, M.: Simulated quantum computation of molecular energies. Science 309(5741), 1704–1707 (2005)
Barenco, A. , Bennett, C.H., Cleve, R., Divincenzo, D.P., Margolus, N., Shor, P., Sleator, T., Smolin, J.A., Weinfurter, H.: Elementary gates for quantum computation. Phys. Rev. A 52(5), 3457–3467 (1995)
Barends, R., Lamata, L., Kelly, J., García-Álvarez, L., Fowler, A.G., Megrant, A., Jeffrey, E., White, T.C., Sank, D., Mutus, J.Y., et al.: Digital quantum simulation of fermionic models with a superconducting circuit. Nat. Commun. 6(7654) (2015)
Barends, R., Shabani, A., Lamata, L., Kelly, J., Mezzacapo, A., Las Heras, U., Babbush, R., Fowler, A.G., Campbell, B., Chen, Y., et al.: Digitized adiabatic quantum computing with a superconducting circuit. Nature 534 (7606), 222–226 (2016)
Benenti, G., Strini, G.: Quantum simulation of the single-particle Schroedinger equation. Am. J. Phys. 76(7), 657–662 (2008)
Bergholm, V., Vartiainen, J.J., Moettoenen, M., Salomaa, M.M.: Quantum circuits with uniformly controlled one-qubit gates. Phys. Rev. A 71, 052330 (2005)
Berry, D.W.: High-order quantum algorithm for solving linear differential equations. J. Phys. A Math. Theor. 47(10), 105301 (2014)
Berry, D.W., Ahokas, G., Cleve, R., Sanders, B.C.: Efficient quantum algorithms for simulating sparse Hamiltonians. Commun. Math. Phys. 270 (2), 359–371 (2007)
Blass, A., Gurevich, Y.: Ancilla-approximable quantum state transformations. J. Math. Phys. 56(4), 042201 (2015)
Boghosian, B.M., Taylor, W.: Simulating quantum mechanics on a quantum computer. Physica D: Nonlinear Phenomena 120(1), 30–42 (1998)
Brown, K.L., Munro, W.J., Kendon, V.M.: Using quantum computers for quantum simulation. Entropy 12(11), 2268 (2010)
Cao, Y., Papageorgiou, A., Petras, I., Traub, J., Kais, S.: Quantum algorithm and circuit design solving the Poisson equation. New J. Phys. 15(1), 013021 (2013)
Cramer, M., Plenio, M.B., Flammia, S.T., Somma, R., Gross, D., Bartlett, S.D., Landon-Cardinal, O., Poulin, D., Liu, Y.-K.: Efficient quantum state tomography. Nat. Commun. 1, 149 (2010)
D’Ariano, G.M., Paris, M.G.A., Sacchi, M.F.: Quantum tomography. Advances in Imaging and Electron Physics 128, 206–309 (2003)
Deutsch, D.: Quantum theory, the Church-Turing principle and the universal quantum computer. Proceedings of the Royal Society of London A: Mathematical, Physical and Engineering Sciences 400(1818), 97–117 (1985)
Douglas, B.L., Wang, J.B.: Efficient quantum circuit implementation of quantum walks. Phys. Rev. A 79, 052335 (2009)
Feynman, R.P.: Simulating physics with computers. Int. J. Theor. Phys. 21 (6), 467–488 (1982)
Fillion-Gourdeau, F., Lorin, E., Bandrauk, A.D.: Numerical solution of the time-dependent Dirac equation in coordinate space without fermion-doubling. Comput. Phys. Comm. 183(7), 1403–1415 (2012)
Fillion-Gourdeau, F., Lorin, E., Bandrauk, A.D.: Resonantly enhanced pair production in a simple diatomic model. Phys. Rev. Lett. 110(1), 013002 (2013)
Fillion-Gourdeau, F., MacLean, S., Laflamme, R.: Algorithm for the solution of the dirac equation on digital quantum computers. Phys. Rev. A 95, 042343 (2017)
Georgescu, I.M., Ashhab, S., Nori, F.: Quantum simulation. Rev. Mod. Phys. 86, 153–185 (2014)
Godlewski, E., Raviart, P.-A.: Hyperbolic Systems of Conservation Laws, vol. 3/4 of mathématiques & Applications (Paris) [Mathematics and Applications]. Ellipses, Paris (1991)
Godlewski, E., Raviart, P.-A.: Numerical Approximation of Hyperbolic Systems of Conservation Laws, vol. 118 of Applied Mathematical Sciences. Springer, New York (1996)
Green, A.S., Lumsdaine, P.L., Ross, N.J., Selinger, P., Valiron, B.: An introduction to quantum programming in quipper. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 7948 LNCS: 110–124 (2013)
Green, A.S., Lumsdaine, P.L., Ross, N.J., Selinger, P., Valiron, B.: Quipper: A scalable quantum programming language. In: Proceedings of the ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI), pp. 333–342 (2013)
Grover, L., Rudolph, T.: Creating superpositions that correspond to efficiently integrable probability distributions. arXiv:quant-ph/0208112 (2002)
Harrow, A.W., Hassidim, A., Lloyd, S.: Quantum algorithm for linear systems of equations. Phys. Rev. Lett. 103(15), 150502,4 (2009)
Jordan, S.P., Lee, K.S.M., Preskill, John: Quantum algorithms for quantum field theories. Science 336(6085), 1130–1133 (2012)
Kassal, I., Jordan, S.P., Love, P.J., Mohseni, M., Aspuru-Guzik, A.: Polynomial-time quantum algorithm for the simulation of chemical dynamics. Proc. Natl. Acad. Sci. 105(48), 18681–18686 (2008)
Kassal, I., Whitfield, J.D., Perdomo-Ortiz, A., Yung, M.-H., Aspuru-Guzik, A.: Simulating chemistry using quantum computers. Annu. Rev. Phys. Chem. 62, 185207 (2011)
Kaye, P., Mosca, M.: Quantum networks for generating arbitrary quantum states. arXiv:quant-ph/0407102 quant-ph/0407102(2004)
Julian Kelly, R., Barends, A.G., Fowler, A., Megrant, E., Jeffrey, T.C., White, D., Sank, J.Y., Mutus, B., Campbell, Y, et al.: Chen State preservation by repetitive error detection in a superconducting quantum circuit. Nature 519(7541), 66–69 (2015)
Labbé, S., Lorin, E.: On the reservoir technique convergence for nonlinear hyperbolic conservation laws. I. J. Math. Anal. Appl. 356(2), 477–497 (2009)
Lanyon, B.P., Hempel, C., Nigg, D., Müller, M., Gerritsma, R., Zähringer, F., Schindler, P., Barreiro, J.T., Rambach, M., Kirchmair, G., Hennrich, M., Zoller, P., Blatt, R., Roos, C.F.: Universal digital quantum simulation with trapped ions. Science 334(6052), 57–61 (2011)
Leveque, R.J.: Finite Volume Methods for Hyperbolic Problems, vol. 31. Cambridge University Press, Cambridge (2002)
Lloyd, S.: Universal quantum simulators. Science 273, 1073–1078 (1996)
Meyer, D.A.: Quantum computing classical physics. Philosophical Transactions of the Royal Society of London A: Mathematical, Phys. Eng. Sci. 360(1792), 395–405 (2002)
Mezzacapo, A., Sanz, M., Lamata, L., Egusquiza, I.L., Succi, S., Solano, E.: Quantum simulator for transport phenomena in fluid flows. Sci. Rep. 5(13153) (2015)
Negrevergne, C., Mahesh, T.S., Ryan, C.A., Ditty, M., Cyr-Racine, F., Power, W., Boulant, N., Havel, T., Cory, D.G., Laflamme, R.: Benchmarking quantum control methods on a 12-qubit system. Phys. Rev. Lett. 96, 170501 (2006)
Nielsen, M.A, Chuang, I.L.: Quantum computation and quantum information. Cambridge University Press, Cambridge (2010)
Papageorgiou, A., Traub, J.F.: Measures of quantum computing speedup. Phys. Rev. A 88, 022316 (2013)
Rønnow, T.F., Wang, Z., Job, J., Boixo, S., Isakov, S.V., Wecker, D., Martinis, J.M., Lidar, D.A., Troyer, M.: Defining and detecting quantum speedup. Science 345(6195), 420–424 (2014)
Salathé, Y., Mondal, M., Oppliger, M., Heinsoo, J., Kurpiers, P., Potočnik, A., Mezzacapo, A., Las Heras, U., Lamata, U., Solano, E., Filipp, S., Wallraff, A.: Digital quantum simulation of spin models with circuit quantum electrodynamics. Phys. Rev. X 5, 021027 (2015)
Serre, D.: Systémes de lois de conservation. I. Fondations. [Foundations]. Diderot Editeur, Paris. Hyperbolicité, entropies, ondes de choc. [Hyperbolicity, entropies, shock waves] (1996)
Shor, P.: Polynomial-time algorithms for prime factorization and discrete logarithms on a quantum computer. SIAM J. Comput. 26(5), 1484–1509 (1997)
Sinha, S., Russer, P.: Quantum computing algorithm for electromagnetic field simulation. Quantum Inf. Process 9(3), 385–404 (2010)
Smoller, J.: Shock waves and reaction-diffusion equations, vol. 258 of Grundlehren der Mathematischen Wissenschaften [Fundamental Principles of Mathematical Science]. Springer, New York-Berlin (1983)
Somma, R., Ortiz, G., Gubernatis, J.E., Knill, E., Laflamme, R.: Simulating physical phenomena by quantum networks. Phys. Rev. A 65, 042323 (2002)
Steane, A.: Quantum computing. Rep. Prog. Phys. 61(2), 117 (1998)
Strikwerda, J.C.: Finite Difference Schemes and Partial Differential Equations, 2nd edn. Society for Industrial and Applied Mathematics (SIAM), Philadelphia (2004)
Vartiainen, J.J., Moetioenen, M., Salomaa, M.M.: Efficient decomposition of quantum gates. Phys. Rev. Lett. 92(17), 177902–1 (2004)
Wang, Xi-Lin, Chen, Luo-Kan, Li, W., Huang, H.-L., Liu, C., Chen, C., Luo, Y.-H., Su, Z.-E., Wu, D., Li, Z.-D., Lu, H., Hu, Y., Jiang, X., Peng, C.-Z., Li, L., Liu, N.-L., Chen, Y.-A., Lu, C.-Y., Pan, J.-W.: Experimental ten-photon entanglement. Phys. Rev. Lett. 117, 210502 (2016)
Wiebe, N., Berry, D., Hoyer, P., Sanders, B.C.: Higher order decompositions of ordered operator exponentials. J. Phys. A Math. Theor. 43(6), 065203 (2010)
Wiesner, S.: Simulations of many-body quantum systems by a quantum computer. arXiv:quant-ph/9603028 quant-ph/9603028
Yung, Man-Hong, Nagaj, Daniel, Whitfield, James D., Aspuru-Guzik, A.: Simulation of classical thermal states on a quantum computer A transfer-matrix approach. Phys. Rev. A 82, 060302 (2010)
Yung, M.-H., Whitfield, J.D., Boixo, S., Tempel, D.G., Aspuru-Guzik, A.: Introduction to Quantum Algorithms for Physics and Chemistry, pp 67–106. Wiley, Hoboken (2014)
Zalka, C.: Efficient simulation of quantum systems by quantum computers. Fortschritte der Physik 46(6-8), 877–879 (1998)
Zalka, C.: Simulating quantum systems on a quantum computer. Proceedings of the Royal Society of London A: Mathematical, Phys. Eng. Sci. 454(1969), 313–322 (1998)
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix A: Numerical example for the reservoir method with diagonal system
We now illustrate this approach with a simple classical diagonal two-dimensional test, with \(\lambda _{1}^{(1)}= 1,\lambda _{2}^{(1)}= 4,\lambda _{3}^{(1)}= 8\) and \(\lambda _{1}^{(2)}= 1,\lambda _{2}^{(2)}= 2,\lambda _{3}^{(2)}= 4\). The computational domain is [0, 10]2, and T = 0.75, m = 3, d = 2, Δx1 = Δx2 = 0.1. The time step is given by Δtn = 0.012488 for all n ≤ nT = 60. The initial data is U0;1(x1,x2) = exp (− 2 ((x1 − 5/2)2 + (x2 − 5/2)2)), U0;2(x1,x2) = exp (− 4 ((x1 − 5/2)2 + (x2 − 5/2)2)), U0;3(x1,x2) = exp (− 8 ((x1 − 5/2)2 + (x2 − 5/2)2)). The reservoir solution components are represented in Fig. ?? (2nd column), showing no numerical diffusion whatsoever, unlike the “CFL= 1” solutions also represented in Fig. ?? (3rd column).
Appendix B: Example for a rotation operator
A simple explicit example for a rotation operator can easily be constructed, inspired from [27]. We consider for d = 3, m = 22, the following system:
with A(x) = S(x)Λ(x)S(xT) (resp. A(y) = S(y)Λ(y)S(y)T, A(z) = S(z)Λ(z)S(z)T), where S(γ) with γ = x,y,z, are defined by
and where {σγ}γ = x,y,z are the Pauli matrices. From a quantum algorithm point of view, these rotations operators can be decomposed [27] as:
where C(σγ) (resp. H) is σγ-controlled (resp. Hadamard) gate (see Fig. 1 in [27]). In this case, the construction of the quantum circuit can directly be deduced from [27], thanks to a simple decomposition in elementary quantum gates of S(γ). More elaborated cases corresponding to more complex S(γ) can be considered as discussed above, but then necessitate more complex quantum circuits.
Appendix C: Determining minimal resource requirements of the quantum algorithm with Quipper
Quipper is a Haskell-based embedded functional language whose purpose is to emulate the implementation of quantum algorithms on realistic quantum computers by providing explicit gate decompositions of quantum algorithms [31, 32]. In particular, functions for transforming complex quantum circuits into elementary gates (Hadamard, CNOT, Clifford...) are included in Quipper, along with many other functionalities allowing for circuit assembly and for resource requirement diagnosis. In this section, we will implement some of the algorithms presented above.
Throughout, it is assumed that all logical operations are implemented without error. In a real quantum device, the quantum system interacts with its environment, generating some noise and error in each operation. In this sense, the gate counts given below represent a lower bound estimates for the “true” algorithm, which may require error-correcting steps.
The algorithm for the solution of hyperbolic systems was formulated in the abstract Hilbert space of the quantum register. Therefore, it is independent of the computer architecture and thus is amenable to any digital quantum devices. For example, quantum computers based on superconducting circuits [10, 11, 39], trapped ions [41], and cavity quantum electrodynamics [50] have been used with some success for other problems and could be used in principle to implement our algorithm. The main limitations however are (i) the number of available qubits in the current register and (ii) the coherence time of these devices that restricts the number of logic quantum gates. State-of-the-art quantum computations on actual digital computers reach ≈ 1000 quantum logic gates on ≈ 9 qubits [11].
In all the examples considered in the following, minimal requirements are studied in order to assess the feasibility of simulations on these real quantum devices. In particular, we count the number of quantum gates (circuit depth) and the total number of qubits (circuit width) required to evolve the initial condition data to a given final time. Due to physical limitations in terms of the number of qubits and the coherence time, only systems with overly small meshes are investigated. As emphasized in [27], this may be enough for proof-of-principle calculations but is far from outperforming classical computations. Nevertheless, given the amount of effort and resources devoted to the development of these quantum devices, these numbers will likely be improved in the future.
1.1 C.1 One-dimensional hyperbolic system
We consider a one-dimensional system with a computational domain as [0, 1] × [0,T]:
where A = Λ is a diagonal matrix in \(M_{2}({\mathbb {R}})\), with eigenvalues λ1 = 1,λ2 = − 3. If A were not diagonal, it would be necessary to add a transition operator at the beginning and the end of the circuit. For T = 10− 1 and nT = 40, we can determine the sets:
We then implement the quantum algorithm described in Section ??, with Quipper. In this case, it is only necessary to implement the decrement and increment operators, thanks to the preliminarily established list \(\mathcal {I}^{n_{T}}\) of characteristics to be updated. Following Table I from [27], we report the minimal requirements for simulating the one-dimensional system described above, for proof-of-principle calculations. More specifically using Quipper, we can evaluate precisely the number of elementary gates necessary to implement the quantum algorithm. Say for respectively nx = 2, nx = 4, and nx = 8, the circuit depth computed by Quipper is 780, 5520, and 27960, and the circuit width is respectively 3, 7, and 15. For instance for nx = 4 that is Nx = 16, the respective number of Hadamard, Clifford, Toffoli, and CNOT gates is found to be 540, 1890, 1350, and 1560. We report in Fig. 10 the quantum circuit for the first 4 iterations with Nx = 16. This quantum circuit has a width equal to 5, that is 4 qubits for labeling the coordinate space positions, and 1 qubit for the labeling of the component. The circuits which are represented correspond to the first 4 time iterations. We notice that the circuit for the first component (top) has 4 times the same pattern (elementary circuit), corresponding to 4 translations from the left to the right, while the circuit for the second component (bottom) corresponds to only 1 iteration from the right to the left, for the same lapse of time. This is due to the fact that λ2 = − 3 × λ1, so that after 4 iterations, 4 translations to the right are applied to the first component, while the second component is only translated once to the left. The quantum circuit for 1 iteration is generated by Quipper.
As an illustration for the same test, but with Nx = 4, we also report in Fig. 11 the quantum circuit for the corresponding gate decomposition and which is still generated by Quipper. This time we only have 3 qubits, 1 for the component index, and 2 for the positions.
The case with the smallest circuit depth, for the lower number of lattice points (nx = 2,Nx = 4), could possibly be implemented on actual device for a proof-of-principle calculation, as long as the initialization phase requires less than approximately 300 gates. However, the results obtained from the gate decomposition demonstrate that it would be a challenging task to perform quantum simulations of hyperbolic systems on actual quantum device with larger lattice size. Moreover, performing a relevant quantum simulation outperforming classical computation demands much improvement from both the coherence time and the number of qubits.
1.2 C.2 Three-dimensional hyperbolic system
In this subsection, we implement with Quipper the quantum version of the reservoir method for a three-dimensional linear hyperbolic system (1) with \(\lambda _{1}^{(\mathrm {x})}=\lambda _{2}^{(\mathrm {x})}=-1\), \(\lambda _{3}^{(\mathrm {x})}=\lambda _{4}^{(\mathrm {x})}=\sqrt {2}\), \(\lambda _{1}^{(\mathrm {y})}=\lambda _{2}^{(\mathrm {y})}=-2\), \(\lambda _{3}^{(\mathrm {y})}=\lambda _{4}^{(\mathrm {y})}= 2\sqrt {2}\), \(\lambda _{1}^{(\mathrm {z})}=\lambda _{2}^{(\mathrm {z})}=-4\), \(\lambda _{3}^{(\mathrm {z})}=\lambda _{4}^{(\mathrm {z})}= 4\sqrt {2}\). Notice that we have associated the following upper indices: (x) ⇔ (1), (y) ⇔ (2), (z) ⇔ (3). We then select S(γ) as in (2), where we recall that Pauli’s matrices are defined by:
The motivation for considering such “simple” transition matrices is to design simple quantum circuit portion for diagonalization. As mentioned in Section ??, more complex transition unitary matrices could be considered, but would then require additional work for decomposing in elementary quantum gates. The quantum algorithm is applied from time 0 to 10− 2, with Δx = 10− 2, corresponding to nT = 30 iterations. In addition, \(\mathcal {I}^{n_{T}}\) and \(\mathcal {S}^{n_{T}}\) are such that:
The corresponding time steps take the value 0.0177, 0.0073, 0.0104, 0.0030, 0.0177, 0.0043⋯. We then encode \(\mathcal {I}^{n_{T}}\) in the quantum algorithm in order to implement the quantum reservoir method. This list provides the directional characteristic field to update. Say for respectively nx = ny = nz = 2, nx = ny = nz = 4, and nx = ny = nz = 8, the circuit depth computed by Quipper is 672, 3042, and 14262, and the circuit width is respectively 8, 16, and 32. For instance for nx = ny = nz = 4 that is Nx = Ny = Nz = 16, the respective numbers of Hadamard, Clifford, Toffoli, and CNOT gates are found to be 400, 1037, 675, and 840. We report in Fig. 12 the quantum circuit for the first iteration with Nx = Ny = Nz = 4, and a circuit width equal to 8. The quantum circuit which is generated by Quipper is much more complex (Toffoli, Hadamard, Clifford quantum gates, etc.), as it also includes the changes of basis (rotations) and the translations.
The same conclusion as in the 1-D case can be reached from these results, i.e., a proof-of-principle calculation could possibly be performed with the smaller systems, but relevant calculations would require improvements in quantum technologies.
Rights and permissions
About this article
Cite this article
Fillion-Gourdeau, F., Lorin, E. Simple digital quantum algorithm for symmetric first-order linear hyperbolic systems. Numer Algor 82, 1009–1045 (2019). https://doi.org/10.1007/s11075-018-0639-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11075-018-0639-3