Pith. sign in

REVIEW 3 major objections 5 minor 2 cited by

Fast and Flexible Quantum-Inspired Differential Equation Solvers with Data Integration

T0 review · 3 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Quantized tensor trains solve linear and nonlinear PDEs with cost and memory that scale logarithmically in the number of space-time grid points.

desk verdict Solid empirical QTT-PDE paper with a genuinely useful data-integration trick; the headline O(log(NT)) claim is not proven, but the method and benchmarks deserve referee time. read the letter →

arxiv 2505.17046 v1 pith:FWR2GPAB submitted 2025-05-15 math.NA cs.LGcs.NAquant-ph

Lucas Arenstein , Martin Mikkelsen , Michael Kastoryano This is my paper · ORCID
classification math.NAcs.LGcs.NAquant-ph MSC 65M0665N0615A6965Y20
keywords quantizedtensortrainsQTTpartialdifferentialequationslogarithmiccomplexityspace-timeformulationdata-drivenPDEsolvingsplineinterpolationBurgersequation
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that a quantum-inspired tensor format, the quantized tensor train (QTT), can serve as a general-purpose PDE solver whose computational cost and memory grow only logarithmically in the number of spatial and temporal grid points. It claims this $O(\log(NT))$ scaling for both linear and nonlinear equations by discretizing space and time together and solving the whole space-time system at once instead of stepping through time. If true, very fine grids that would be out of reach for finite-difference or multigrid methods become inexpensive, and the Courant-Friedrichs-Lewy stability condition that limits explicit time-stepping is sidestepped. The paper also reports a data-driven extension: boundary and source data are learned from samples through spline fitting followed by QTT interpolation, giving fast and accurate solvers without a separate neural-network training phase.

What carries the argument

The quantized tensor train (QTT) is a tensor network in which a function on a $2^c$-point grid is written as a product of $c$ small cores, with each index carrying one binary digit of the coordinate so that the network encodes length scales hierarchically. The load-bearing facts are explicit low-rank QTT representations of the standard finite-difference operators, in particular a fixed rank-3 representation of any tridiagonal Toeplitz matrix (Lemma 1), and of smooth elementary functions such as sine, exponential, and polynomials, which keep operator and source ranks constant as the grid is refined. The (modified) alternating linear scheme (M)ALS then solves the resulting linear systems by sweeping through the cores one or two at a time, acting as an efficient preconditioned iterative solver restricted to the manifold of bounded-rank QTT states.

What would settle it

Take a PDE with a solution that develops a sharp moving front or shock, such as small-viscosity Burgers' equation, and solve it on grids from $2^8$ to $2^{16}$ points while recording the internal QTT rank of the recovered solution and time-to-accuracy. If the rank grows with grid resolution, or if the runtime per accuracy bends upward from the logarithmic trend, then the claimed $O(\log(NT))$ complexity fails for that regime.

Watch

Extended reading notes

Core claim

The central claim is that discrete differential operators and PDE solutions can be compressed into QTT format with bond dimension independent of grid resolution, and that an alternating linear scheme (ALS) solver on the QTT system then runs in $O(c r^3 R^2 n^2)$, which with bounded ranks $r,R$ and mode size $n=2$ is logarithmic in the $2^c$ grid size. The paper demonstrates this on 2D and 3D Poisson problems and on Burgers' equation, including a space-time formulation where time is treated as a spatial coordinate, reaching an overall complexity of $O(\log(NT))$ for the first time. To bring data-driven boundary conditions and source terms into the format, the authors fit splines to data and use a multiscale interpolative QTT construction, reporting solver runtimes orders of magnitude faster than physics-informed neural networks at comparable accuracy.

Load-bearing premise

The method's promised speed rests on one load-bearing premise: a fine-grid solution can still be stored in the compressed QTT format without needing appreciably more internal data as the grid gets finer, and the solver's fixed number of correction passes still converges at that fine resolution.

Editorial extensions

If this is right

  • In the paper's 3D anisotropic Poisson benchmark, the QTT solver reaches MSE $2\times10^{-13}$ in about 0.03 seconds while an algebraic-multigrid solver needs 347 seconds for the same accuracy.
  • The space-time formulation solves the full space-time solution at once; time-stepping is replaced by a few outer iterations ('runs') for the nonlinear term, and the Courant-Friedrichs-Lewy condition no longer restricts the time step.
  • The Burgers' equation space-time solver reports reaching MSE around $10^{-22}$ for a very small viscosity on a $2^{14}\times2^{14}$ grid in about 0.01 seconds, a regime the paper says is unattainable with the time-stepping version.
  • Data-driven boundary and source terms learned from samples via spline fitting and QTT interpolation enter the solver with only a small runtime overhead, and the paper reports reaching accuracy comparable to a physics-informed neural network in under a second rather than hundreds of seconds.
  • Because the method treats time as a spatial dimension, a single solve yields the full space-time solution, enabling parallel-in-time analysis that sequential time-stepping cannot offer.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A direct extension not explored in the paper is to apply the spline-to-QTT interpolation to inverse problems where observations are scattered inside the domain, not only on the boundary; the same pipeline that learns boundary data should learn source terms from interior samples.
  • The paper's fixed-number-of-sweeps recipe suggests an adaptive-rank variant: grow the QTT bond dimension only as needed based on residuals, which would make the method robust to problems where the assumed rank bound is violated.
  • If the logarithmic scaling survives nonlinear and chaotic regimes, the space-time QTT solver becomes a candidate for global-in-time sensitivity analysis and uncertainty quantification, since the whole solution is available at once rather than as a sequence of time steps.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The manuscript presents a quantized tensor train (QTT) framework for solving linear and nonlinear PDEs, with applications to 2D/3D Poisson equations, Burgers' equation, and heat-type problems. The authors propose finite-difference discretizations in QTT format, an alternating least squares (ALS/MALS) linear solver, and a space-time formulation that treats time as an additional spatial dimension, claiming for the first time an overall complexity of O(log(NT)) in the number of spatial and temporal degrees of freedom. They additionally introduce a spline-interpolation-based data integration technique for learning boundary and source terms from data, and compare their solvers against PyAMG and PINN baselines on several benchmark problems.

Significance. If the central complexity claim is correct, the paper would make a useful contribution: QTT-based space-time solvers with logarithmic memory and runtime scaling would be genuinely valuable for high-resolution and high-dimensional PDEs. The manuscript has several strengths: explicit analytic QTT constructions (Lemma 1 and Appendix A), a reproducible code release, and numerical comparisons against established classical (AMG) and learning-based (PINN) solvers. The space-time treatment avoiding CFL restrictions is also conceptually appealing. However, the logarithmic scaling is supported only by empirical tables, with no a priori rank bounds or convergence theory, and at least one reported accuracy value is below what double-precision arithmetic can plausibly certify. The significance of the work is therefore currently conditional on filling these gaps.

major comments (3)
  1. [Section I, paragraph 5; Section II F; Section IV C] The central claim of O(log(NT)) complexity requires that the QTT bond dimension of the solution and the number of MALS sweeps and Picard runs remain bounded as N and T are refined. The paper does not prove this for any of the PDEs considered, and it does not report the observed bond dimensions as a function of grid size. Since the ALS complexity in Section II F is O(c gamma r^3 R^2 n^2), any growth of r with refinement would destroy the logarithmic claim. The statement in Section IV C that 'the number of runs does not need to be increased with the grid resolution' is an empirical observation for one benchmark, not a general fact; the paper should either prove rank-uniformity results for the model problems or substantially weaken the claim to 'logarithmic scaling observed on low-rank benchmarks' and report rank trajectories alongside runtime.
  2. [Table II, last row] The entry '1e-07 1.05 - - 0.0116 2.12e-22' reports an MSE of 2.12e-22 for a Float64 computation on an M3 processor. This is far below machine epsilon for quantities of unit scale, and it is not credible unless the comparison is made on a very small subset of grid points or the analytical solution is evaluated in extended precision. The authors should state exactly how the MSE is computed (which grid points, which evaluation formula, what precision), and either correct this value or provide a reproducibility experiment. Since this entry is part of the scaling evidence in the table, the inconsistency weakens the numerical support for the accuracy claims.
  3. [Section VI; Section II C] The paper explicitly defers to future work the development of error bounds and stability guarantees for nonlinear operators, and it asserts without proof that the QTT ranks of the discrete solutions remain low. In a numerical analysis journal, the absence of any convergence or rank-growth analysis is a load-bearing gap for a paper whose headline contribution is a complexity guarantee. I recommend adding (i) a formal statement of the assumptions under which O(log(NT)) holds, (ii) numerical rank monitoring on refinement paths for the Burgers benchmarks, including small viscosity (nu = 1e-5, 1e-7) and the sharp-kink Poisson problem, and (iii) at least a heuristic argument for why MALS with two sweeps and two runs suffice independently of grid size.
minor comments (5)
  1. [Section II A] There is a grammatical error in 'An tensor T is a d-dimensional array'; it should be 'A tensor'.
  2. [Section II F] The complexity expression O(c gamma r^3 R^2 n^2) uses symbols r and R that are not formally defined in the equation; please state explicitly that r is the maximum QTT rank of the iterates and R is the maximum rank of A_QTT, and specify how ranks are truncated during ALS sweeps.
  3. [Algorithm 5, line 5] The line 'w_sol[k+1] ← ALS(A_QTT, w_sol[k], w_sol[k] + (4))' is unclear: the '(4)' appears to refer to a vector constructed in a previous step, but this is not defined in the algorithm box; please rephrase.
  4. [Appendix A.1] In the boundary-vector construction, the same component is listed twice ('F 0,0,0,1 = v_a, F 0,0,0,1 = v_b'); the second should presumably have a different index to place the boundary value at the opposite end.
  5. [Section V, Table III] The two columns 'Best<1 sec' and 'MSE<e-04' presumably correspond to different solver settings (MALS vs ALS) but the hyperparameters M and sweep counts are not reported; please provide the exact configuration for each column so the comparison is reproducible.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the O(log(NT)) claim is a conditional complexity statement, benchmarks are external, and no fitted parameter is renamed as a prediction.

full rationale

Walking the derivation chain, the only load-bearing complexity statement is conditional. In Section II F the paper states: 'The computational complexity of ALS is O(cγr^3R^2n^2), where c is the number of cores...' and then 'since A_QTT admits an exact low-rank representation and we often also obtain a low-rank representation for b_QTT, the overall complexity can remain polynomial in c'. The abstract's and Section I's 'O(log(NT))' is simply this formula with c = log2 N and bounded ranks/sweeps; the paper does not derive the bound from the target MSE or from a self-citation. The numerical claims are tested against analytical solutions (Poisson and Burgers, Tables I-III and C) and external solvers (PyAMG, PINN benchmarks in [36]); no parameter is fitted to the reported MSE and then relabeled as a prediction. The data-driven section's hyperparameter M is chosen by hand and the PDE solution is a genuine forward solve. The cited QTT constructions [17,28,32,33] are external prior work, not the present authors' own citations, so there is no self-citation chain. Caveats such as the sub-roundoff MSE of 2.12e-22 in Table II and the unproved uniformity of ranks/sweeps are correctness risks, not circularity.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The free parameters are numerical method hyperparameters, not fitted unknowns. The central complexity claim rests on the unproven assumption that solution QTT ranks and ALS sweep counts stay bounded as the grid refines. The paper provides no proof of this, and the data-driven construction explicitly introduces a user-chosen bond dimension via M.

free parameters (4)
  • M (number of interpolation nodes) = Not fitted; user-chosen, e.g., 25 in Figure 3
    In Algorithm 3 and Section II D d, M determines the QTT bond dimension (M+1). The reported data-driven results depend on M; the paper selects different M values for different accuracy targets in Table III.
  • runs (Picard iterations in space-time solver) = 2 in most reported results; 5 or 7 in Problem 4
    In Algorithm 2, runs controls how many times the nonlinear term is updated. Figure 2 shows accuracy plateaus after 2 runs, but the choice is manual.
  • ALS/MALS sweeps = 2 in benchmark configurations
    Appendix C states MALS with 2 sweeps was used; the runtime scaling to O(log N) depends on this count staying fixed as grid size grows.
  • Initial guess rank structure = Random low-rank QTT; exact rank not specified
    Section II F and Appendix C note the initial guess rank affects accuracy and runtime; variations in runtime across runs are attributed to this randomness.
assumptions (4)
  • domain assumption The discrete solution of the PDE has a low-rank QTT representation with bond dimension independent of grid size.
    Logarithmic complexity claim requires bounded rank; this is only observed in examples, never proven. For the source term learned from data, the rank is tied to the user-chosen M, so it is not independent.
  • domain assumption ALS/MALS converges to a sufficiently accurate solution in a fixed number of sweeps independent of grid size.
    No convergence theorem is provided; runtime scaling O(log N) assumes sweep count does not grow with grid refinement.
  • standard math The multiscale interpolative QTT construction from Lindsey [17] provides accurate representations of spline-fitted data and is stable for noisy data.
    The data-driven section relies on this construction; details are external to the paper and the paper does not prove its accuracy on the presented data.
  • domain assumption The finite-difference schemes used (implicit time-stepping and space-time global solve) are stable and converge to the analytical solution on the chosen grids.
    No stability or convergence analysis is included; only numerical evidence is provided, and Section VI explicitly lists a rigorous convergence theory as future work.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Fast and Flexible Quantum-Inspired Differential Equation Solvers with Data Integration." pith.science (2026). https://pith.science/paper/FWR2GPAB

@misc{pith2026250517046,
  author       = {Pith},
  title        = {Pith review of: Fast and Flexible Quantum-Inspired Differential Equation Solvers with Data Integration},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FWR2GPAB}},
  note         = {Machine review of arXiv:2505.17046}
}
read the original abstract

Accurately solving high-dimensional partial differential equations (PDEs) remains a central challenge in computational mathematics. Traditional numerical methods, while effective in low-dimensional settings or on coarse grids, often struggle to deliver the precision required in practical applications. Recent machine learning-based approaches offer flexibility but frequently fall short in terms of accuracy and reliability, particularly in industrial contexts. In this work, we explore a quantum-inspired method based on quantized tensor trains (QTT), enabling efficient and accurate solutions to PDEs in a variety of challenging scenarios. Through several representative examples, we demonstrate that the QTT approach can achieve logarithmic scaling in both memory and computational cost for linear and nonlinear PDEs. Additionally, we introduce a novel technique for data-driven learning within the quantum-inspired framework, combining the adaptability of neural networks with enhanced accuracy and reduced training time.

Figures

Figures reproduced from arXiv: 2505.17046 by the authors.

Figure 1
Figure 1. FIG. 1: Log-log plot comparing the run time and [PITH_FULL_IMAGE:figures/full_fig_p006_1.png] view at source ↗
Figure 2
Figure 2. FIG. 2: MSE of the space-time QTT solver as a function [PITH_FULL_IMAGE:figures/full_fig_p010_2.png] view at source ↗
Figure 3
Figure 3. FIG. 3: QTT interpolation on 7 data points sampled [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: FIG. 4: Point-wise absolute error between the QTT and [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: FIG. 5: Solution of the 2D heat equation at different time steps. [PITH_FULL_IMAGE:figures/full_fig_p018_5.png]
Figure 6
Figure 6. Figure 6: FIG. 6: Comparison of QTT solutions (top row) and slices (bottom row) for different numbers of [PITH_FULL_IMAGE:figures/full_fig_p022_6.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Simulating Quantum Turbulence with Matrix Product States

    quant-ph 2025-08 conditional novelty 6.0 of 10

    A quantics matrix product state solver for the Gross-Pitaevskii equation reproduces soliton, vortex, and quantum turbulence dynamics with 10x to 10,000x memory reduction relative to direct numerical simulation.

  2. Tensor-network approach to quantum optical state evolution beyond the Fock basis

    quant-ph 2025-11 conditional novelty 5.0 of 10

    A tensor-network (MPS/MPO) solver simulates SPDC quantum dynamics directly in the continuous quadrature representation, compressing the state >3,000× at α=100.

Reference graph

Works this paper leans on

47 extracted references · 33 canonical work pages · cited by 2 Pith papers

  1. [36]

    Z. Hao, J. Yao, C. Su, H. Su, Z. Wang, F. Lu, Z. Xia, Y. Zhang, S. Liu, L. Lu,et al., arXiv preprint arXiv:2306.08827 (2023)

    arXiv 2023

  2. [1]

    Brandt, Mathematics of computation31, 333 (1977)

    A. Brandt, Mathematics of computation31, 333 (1977)

    work page 1977

  3. [2]

    Trottenberg, C

    U. Trottenberg, C. W. Oosterlee, and A. Schuller,Multi- grid methods(Academic press, 2001)

    work page 2001

  4. [3]

    Gottlieb and S

    D. Gottlieb and S. A. Orszag,Numerical analysis of spec- tral methods: theory and applications(SIAM, 1977)

    work page 1977

  5. [4]

    J. P. Boyd,Chebyshev and Fourier spectral methods (Courier Corporation, 2001)

    work page 2001

  6. [5]

    Raissi, P

    M. Raissi, P. Perdikaris, and G. E. Karniadakis, Journal of Computational physics378, 686 (2019)

    2019

  7. [6]

    Berg and K

    J. Berg and K. Nystr¨ om, Journal of Computational Physics384, 239 (2019)

    work page 2019

  8. [7]

    Z. Long, Y. Lu, and B. Dong, Journal of Computational Physics399, 108925 (2019)

    work page 2019

Show all 47 references
  1. [8]

    G. Pang, M. D’Elia, M. Parks, and G. E. Karniadakis, Journal of Computational Physics422, 109760 (2020)

    2020

  2. [9]

    L. Lu, P. Jin, G. Pang, Z. Zhang, and G. E. Karniadakis, Nature machine intelligence3, 218 (2021)

    2021

  3. [10]

    Z. Li, N. Kovachki, K. Azizzadenesheli, B. Liu, A. Stu- art, K. Bhattacharya, and A. Anandkumar, Advances in Neural Information Processing Systems33, 6755 (2020)

    2020

  4. [11]

    S. R. White, Phys. Rev. Lett.69, 2863 (1992)

    1992

  5. [12]

    I. V. Oseledets, Dokl. Math.80, 653 (2009)

    2009

  6. [13]

    I. V. Oseledets, SIAM J. on Matrix Anal. Appl.31, 2130 (2010)

    2010

  7. [14]

    Oseledets and E

    I. Oseledets and E. Tyrtyshnikov, Linear Algebra Appl. 432, 70 (2010)

    2010

  8. [15]

    I. V. Oseledets, SIAM J. Sci. Comput.33, 2295 (2011)

    2011

  9. [16]

    M. K. Ritter, Y. N´ u˜ nez Fern´ andez, M. Wallerberger, J. von Delft, H. Shinaoka, and X. Waintal, Phys. Rev. Lett.132, 056501 (2024)

    2024

  10. [17]

    Lindsey, arXiv preprint arXiv:2311.12554 (2023)

    M. Lindsey, arXiv preprint arXiv:2311.12554 (2023)

    2023 arXiv

  11. [18]

    Ye and N

    E. Ye and N. F. Loureiro, Journal of Plasma Physics90, 805900301 (2024)

    2024

  12. [19]

    Gourianov,Exploiting the structure of turbulence with tensor networks, Ph.D

    N. Gourianov,Exploiting the structure of turbulence with tensor networks, Ph.D. thesis, University of Oxford (2022)

    2022

  13. [20]

    Gourianov, M

    N. Gourianov, M. Lubasch, S. Dolgov, Q. Y. van den Berg, H. Babaee, P. Givi, M. Kiffner, and D. Jaksch, Nature Computational Science2, 30 (2022)

    2022

  14. [21]

    Gourianov, P

    N. Gourianov, P. Givi, D. Jaksch, and S. B. Pope, Sci- ence Advances11, eads5990 (2025)

    2025

  15. [22]

    Khoromskij, Constructive Approximation - CONSTR APPROX34(2009), 10.1007/s00365-011-9131-1

    B. Khoromskij, Constructive Approximation - CONSTR APPROX34(2009), 10.1007/s00365-011-9131-1

    2009 doi

  16. [23]

    Richter, L

    L. Richter, L. Sallandt, and N. N¨ usken, inInterna- tional Conference on Machine Learning(PMLR, 2021) pp. 8998–9009

    2021

  17. [24]

    Corona, A

    E. Corona, A. Rahimian, and D. Zorin, Journal of Com- putational Physics334, 145 (2017)

    2017

  18. [25]

    Matrix product state representations,

    D. Perez-Garcia, F. Verstraete, M. M. Wolf, and J. I. Cirac, “Matrix product state representations,” (2007), arXiv:quant-ph/0608197 [quant-ph]

    2007 arXiv

  19. [26]

    Kl¨ umper, A

    A. Kl¨ umper, A. Schadschneider, and J. Zittartz, Euro- physics Letters24, 293 (1993)

    1993

  20. [27]

    I. V. Oseledets, SIAM Journal on Scientific Computing 33, 2295 (2011)

    2011

  21. [28]

    V. A. Kazeev and B. N. Khoromskij, SIAM journal on matrix analysis and applications33, 742 (2012)

    2012

  22. [29]

    Oseledets and E

    I. Oseledets and E. Tyrtyshnikov, Linear Algebra and its Applications432, 70 (2010)

    2010

  23. [30]

    Dolgov and D

    S. Dolgov and D. Savostyanov, Computer Physics Com- munications246, 106869 (2020)

    2020

  24. [31]

    Y. N. Fern´ andez, M. K. Ritter, M. Jeannin, J.-W. Li, T. Kloss, T. Louvet, S. Terasaki, O. Parcol- let, J. von Delft, H. Shinaoka,et al., arXiv preprint arXiv:2407.02454 (2024)

    2024 arXiv

  25. [32]

    Dolgov,Tensor product methods in numerical simula- tion of high-dimensional dynamical problems, Ph.D

    S. Dolgov,Tensor product methods in numerical simula- tion of high-dimensional dynamical problems, Ph.D. the- sis (2014)

    2014

  26. [33]

    Holtz, T

    S. Holtz, T. Rohwedder, and R. Schneider, SIAM Jour- nal on Scientific Computing34, A683 (2012)

    2012

  27. [34]

    Lubasch, P

    M. Lubasch, P. Moinier, and D. Jaksch, Journal of Com- putational Physics372, 587–602 (2018)

    2018

  28. [35]

    Wood, Communications in Numerical Methods in En- gineering22, 797 (2006)

    W. Wood, Communications in Numerical Methods in En- gineering22, 797 (2006)

    2006

  29. [37]

    Burgers’ equation solution using hermite quadrature,

    J. Burkardt, “Burgers’ equation solution using hermite quadrature,” (Accessed 2025)

    2025

  30. [38]

    Basdevant, M

    C. Basdevant, M. Deville, P. Haldenwang, J. Lacroix, J. Ouazzani, R. Peyret, P. Orlandi, and A. Patera, Com- puters & Fluids14, 23 (1986). 13 Appendix A: Useful QTT Constructions In this section, we present key constructions for building QTT representation of matrices which ar...

    1986

  31. [39]

    Boundary Vector

    Build QTT representation of “Boundary Vector” Given theboundary vectorv= va 0 ··· 0 vb ⊺ of length 2c,c≥3 the following construction builds the QTT representation ofvwithccores: v= F 1,2,1,2 ▷◁M1 2,2,1,2 ▷◁(M2 (2,2,1,2))▷◁(c−3)▷◁L (2,2,1,1), with all the entries equal to zero ...

  32. [40]

    Diagonal QTT

    Build “Diagonal QTT” from vector Givenv= v1 v2 ··· v2c ⊺ as input this method builds theccores MPO representation of D= diag(v 1,v 2,...,v 2c). The first step is: v= (v 1 v2 ···v 2c)⊺ MPS ofv Any Method from Section II D V1 V2 Vc Next, we build the following MPO: D1 D2 Dc s.t....

  33. [41]

    Eraser QTT

    Build “Eraser QTT” Given the constantsn 1,n 2,n 3,n 4 andcthis method construct a QTT representation of M a 2 c×2c matrix with all its elements equal to zero except for the four entries shown below: M =   n1 n2 0 0 0 0 n3 n4   (2c×2c) , the construction is given ...

  34. [42]

    The trick of this construction is to use the trigonometric identity sin(α±β) = sin(α) cos(β)±cos(α) sin(β)

    Building the Analytic QTT Representation off(x) = sin(αx+ϕ) We construct an analytic rank-2 QTT representation off(x) = sin(αx+ϕ), wherexis discretized in the interval (0,1) with 2 c grid points. The trick of this construction is to use the trigonometric identity sin(α±β) = si...

  35. [43]

    Time(s) Build B.C

    Building the Analytic QTT Representation off(x) =e αx We construct the analytic rank-1 QTT representation off(x) =e αx, wherexis discretized in the interval (0,1) with 2c grid points. Using our indexing convention, the QTT representation of the discretized functionffunction is...

  36. [44]

    Consider the Laplace equation ∇2u= 0,(x,y)∈(0,1)×(0,1), with boundary conditions: u(x,0) = sin(kπx) sinh(kπ), u(x,1) = 0, u(0,y) = 0, u(1,y) = 0

    Problem 1 - Section III A Here we have the results that were used to build the plot in Section III A and also present a plot of the analytical solution. Consider the Laplace equation ∇2u= 0,(x,y)∈(0,1)×(0,1), with boundary conditions: u(x,0) = sin(kπx) sinh(kπ), u(x,1) = 0, u(...

  37. [45]

    QTT Analytic solver

    Problem 2 - Section III A In Section III A we consider the Poisson equation: ∂2u ∂x2 +ϵ 1 ∂2u ∂y 2 +ϵ 2 ∂2u ∂z 2 =−sin(πx) sin(πy) sin(πz),(x,y,z)∈(0,1) 3, with all zero boundary conditions forϵ 1 = 0.001 andϵ 2 = 0.0001 now we present the results forϵ 1 andϵ 2 = 1. The analyt...

  38. [46]

    QTT Optimized

    Problem 3 For the next problem we will consider other two types of QTT solver. The first solver uses the same default configuration but builds the source term using TT-SVD. The “QTT Optimized” solver has the same configuration as “QTT Interpolation,” but its initial random gue...

  39. [47]

    We consider the equation with the following initial and boundary conditions: ∂u ∂t = (0.01/π)∂2u ∂x2−u ∂u ∂x, x∈[−1,1],t∈[0,1] u(x,0) =−sin(πx), u(−1,t) =u(1,t) = 0

    Problem 4 In this section, we analyze the effect of therunsparameter in the Space-Time QTT Algorithm 2 on the solution of a specific instance of Burgers’ equation. We consider the equation with the following initial and boundary conditions: ∂u ∂t = (0.01/π)∂2u ∂x2−u ∂u ∂x, x∈[...

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.