Pith. sign in

REVIEW 4 major objections 6 minor 76 references

Randomized Tucker-Sketched GMRES

T0 review · 4 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read The paper proposes two randomized Tucker-format GMRES variants that replace full Arnoldi orthogonalization with short recurrences, and it argues that the low-rank Tucker constraint itself acts as an implicit regularizer in inverse problems.

desk verdict Solid, citable solvers with a load-bearing stopping-rule gap; deserves revision, not rejection. read the letter →

arxiv 2608.11091 v1 pith:364A4KOR submitted 2026-08-11 math.NA cs.NA

Alberto Bucci , Martina Iannacito , Mirjeta Pasha , Rudi Smith This is my paper · ORCID
classification math.NAcs.NA MSC 15A6965F1065F5568W2065J2265N2265F22
keywords tensorequationsrandomizednumericallinearalgebraTuckerdecompositionsketchedGMRESKrylovsubspacemethodinverseproblemspreconditioningcompression
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 is about solving linear equations whose unknown is a multidimensional array (a tensor), with operators that are sums of separable terms, as arise in high-dimensional PDEs and imaging inverse problems. Standard Krylov solvers become impractical here because each new basis vector must be stored and orthogonalized against all previous ones, and in low-rank tensor formats the ranks of these basis vectors inflate as iteration proceeds. The paper proposes two randomized solvers in the sketched-GMRES family—one using randomized HOSVD with adaptive rank selection, one using a streamable Nyström approximation with fixed rank—that orthogonalize only against a short window of recent vectors and keep ranks bounded by randomized Tucker compression. The claim is that both beat existing low-rank Tucker solvers in symmetric and nonsymmetric benchmarks, and that for inverse problems the low-rank constraint acts as an implicit regularizer, so adding an adaptive projected Tikhonov term with automatic parameter selection gives stable reconstructions without knowing the noise level.

What carries the argument

The argument runs on the Tucker format, which stores a $d$-dimensional tensor as a small core tensor plus one factor matrix per mode, and on a Khatri-Rao structured Gaussian sketch $\mathbf S = \mathbf S_1 \odot \cdots \odot \mathbf S_d$, defined by folding independent Gaussian matrices $\mathbf S_j$ with $\odot$ denoting the column-wise Kronecker (Khatri-Rao) product. This sketch maps the vectorization of any basis tensor to a short vector without forming dense intermediates, turning GMRES's least-squares problem into the sketched problem $\min_{\mathbf y}\|\mathbf M\mathbf y-\mathbf b_s\|_2$ and permitting a short-recurrence window of width $k_{\mathrm{trunc}}$ in place of full Arnoldi orthogonalization. In the MLN variant the same sketch is produced at no extra cost as the superdiagonal of the Nyström-sketched core, using the identity $[\mathrm{diag}(\mathbf G_\Psi)]_i = (\boldsymbol\psi_{1,i}\otimes\cdots\otimes\boldsymbol\psi_{d,i})^T\mathrm{vec}(\mathbf X)$, where $\otimes$ is the Kronecker product. The role of this machinery is to keep the multilinear ranks of the basis tensors bounded while making the residual computation and stopping test cheap, which is what converts GMRES from an iteration-count-limited method into one whose per-step cost is stable.

What would settle it

Run either proposed solver on the paper's unpreconditioned 3D convection-diffusion benchmark with the orthogonalization window $k_{\mathrm{trunc}}=1$ and the sketch-correction factor $\eta=0.3$; the reported figure already shows true-residual stagnation for RHOSVD-Tucker sGMRES. If increasing the sketch size $s$ while keeping the window fixed still cannot drive the true residual below the prescribed tolerance, then the structured sketch is not a reliable stand-in for the true residual in exactly the nonsymmetric regime the method targets.

Watch

Extended reading notes

Core claim

The central discovery, stated on the paper's own terms, is that the two bottlenecks of Tucker-format GMRES—rank growth of the Krylov basis and the quadratic cost of full Arnoldi orthogonalization—can be attacked simultaneously within the sketched-GMRES framework. RHOSVD-Tucker sGMRES applies randomized HOSVD with an adaptive rank heuristic inside every tensor summation, so basis tensors are rounded without forming dense intermediates, and it orthogonalizes only against the previous $k_{\mathrm{trunc}}$ basis tensors. MLN-Tucker sGMRES uses the multilinear Nyström approximation with a fixed rank, sketches each basis tensor in a single pass, discards the tensor itself, and reconstructs the final solution from the accumulated sketches; the sketch needed for the residual test is obtained for free as the superdiagonal of the Nyström-sketched core. The paper's experiments show these variants reaching the target residual faster than the full-orthogonalization baseline on a preconditioned 3D Poisson equation and an unpreconditioned 3D convection-diffusion equation, and on a 3D image-deblurring inverse problem the paper argues that the low-rank Tucker constraint suppresses high-frequency noise and, together with GCV-selected projected Tikhonov regularization, prevents the semi-convergence that makes unregularized iterates useless.

Load-bearing premise

The load-bearing premise is that the Khatri-Rao structured Gaussian sketch preserves residual norms well enough that minimizing the sketched least-squares problem and stopping on the sketched residual behave like the true residual, a property the paper assumes without proof and that the convection-diffusion experiment shows can fail at $\eta=0.3$.

Editorial extensions

If this is right

  • Per-iteration cost no longer grows with the total iteration count, because only a window of $k_{\mathrm{trunc}}$ recent basis tensors is orthogonalized instead of all previous ones.
  • The MLN variant never needs to store the full Krylov basis: each tensor is sketched once and discarded, and the final solution is assembled from the accumulated sketches, cutting memory from linear-in-iterations growth to a bounded window plus compact sketch storage.
  • In inverse problems, restricting all iterates to a manifold of bounded multilinear rank suppresses high-frequency noise, and the GCV-selected Tikhonov parameter removes the need for oracle-based stopping rules.
  • The same short-recurrence construction works with preconditioning through a flexible-GMRES wrapper, so rank-truncated preconditioner actions can be used without breaking the iteration.
  • The numerical comparison indicates the methods are competitive with symmetric short-recurrence solvers on a preconditioned Poisson problem, not only with other GMRES variants.

Reading between the lines

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

  • Beyond the paper, the two solvers' rank strategies are complementary, suggesting a hybrid that starts with adaptive RHOSVD ranks and switches to a fixed-rank streaming MLN phase once a stable rank is observed.
  • Beyond the paper, the claim that Tucker truncation regularizes inverse problems could be quantified by computing the equivalent filter factors of the rank-truncated iteration and comparing them with classical regularization parameter choices.
  • Beyond the paper, the memory-efficient MLN downsampling introduces a user-set reconstruction resolution $r_{\mathrm{sol}}$; a natural extension is an adaptive rule that grows $r_{\mathrm{sol}}$ until two successive reconstructions agree.
  • Beyond the paper, for operators with slow singular-value decay the fixed-rank MLN approximation will lose accuracy, so updating the target rank adaptively while preserving the streaming structure is a testable extension the paper does not develop.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes two randomized GMRES-type solvers for linear systems with Tucker-structured tensors. RHOSVD-Tucker sGMRES combines a truncated-window partial orthogonalization with randomized HOSVD-based summation and rank selection; MLN-Tucker sGMRES instead uses a fixed-rank multilinear Nyström approximation, so that only compact sketched representations of the Krylov basis are retained. Both methods replace the full Arnoldi orthogonalization with a sketched least-squares problem whose stopping criterion uses the sketched residual. Numerical experiments on a preconditioned 3D Poisson equation, a 3D convection-diffusion problem, and a 3D image deblurring inverse problem compare the methods with RHOSVD-Tucker GMRES and, in the symmetric case, TK-SS-SD. The paper claims that the proposed methods outperform standard low-rank Tucker solvers in time and memory and that the low-rank Tucker constraint acts as an implicit regularizer for inverse problems when combined with adaptive projected Tikhonov penalization and GCV parameter selection.

Significance. The paper addresses a real computational bottleneck: rank growth and full orthogonalization make standard Krylov methods expensive for Tucker-format equations. The algorithmic idea of reusing the MLN sketch, and in particular the identity diag(G_Ψ) = (Ψ_1 ⊙ ⋯ ⊙ Ψ_d)^T vec(Σ c_j X^{(j)}) proved in Section 3.3, is clean and useful, and the streaming memory-efficient variant is a meaningful contribution. The manuscript also ships code, data, and results with a Zenodo DOI, which is commendable. However, the central algorithmic claims currently rest on single runs with many hand-tuned parameters and without a supporting theory for the structured sketch; the paper itself documents a failure of the sketched-residual stopping criterion in Table 2. These issues are load-bearing because the algorithms' output guarantee and the claimed speed/memory advantages depend on the reliability of that stopping rule. If the stopping rule and the structured-sketch embedding question are resolved, the paper would be a solid contribution to randomized tensor solvers.

major comments (4)
  1. [Section 3.2, Algorithm 2 lines 17–19; Algorithm 6 lines 27–29; Table 2] The stopping rule is based on the sketched relative residual ∥M y_k − b_s∥_2/β_s, with termination when it falls below η·tol, but no bound relates this quantity to the true relative residual ∥L(X_k) − B∥_F/∥B∥_F. The paper acknowledges the risk in Section 3.2, and Section 4.2 documents the failure: Table 2 reports RHOSVD-Tucker sGMRES stopping after 97 iterations with true residual 5.09·10^{-5}, above the target 5·10^{-5} stated in the table (and ten times above the 5·10^{-6} target stated in the prose of Section 4.2). The correction η = 0.3 is hand-tuned, and no experiment shows that it transfers across problems, tolerances, or random draws. This is load-bearing because the output guarantee of Algorithms 2 and 6 ('such that ∥L(eX)−B∥_F ≤ tol') is exactly what the stopping rule must certify. Please either provide a certified bound for the structured sketch or modify the algorithms to verify the true residual, for example by occasional exact residual computations, and re-report the experiments accordingly.
  2. [Section 3.2, Definition 4 and Eq. (11)] The reduction to the sketched least-squares problem requires S to be an oblivious subspace embedding for the subspace containing the residual vectors. In this paper S = S_1 ⊙ ⋯ ⊙ S_d with Gaussian S_k (Algorithm 2 lines 3–4; Section 3.3, Eqs. (14) and (21)), but no theorem is given showing that this Khatri–Rao-structured sketch satisfies the (ε, δ, k)-OSE property for the Krylov subspace. The sketch size is chosen by tensor-rank heuristics (RHOSVD) or fixed a priori (MLN), not by the subspace dimension k and the desired distortion ε. The cited analysis in [22] concerns structured sketches for the MLN approximation itself and, as far as the manuscript shows, does not establish a residual-norm subspace embedding for the sGMRES least-squares problem. This gap directly underlies the failure described in the previous comment and should be addressed, even if only with a restricted result for tensors of bounded multilinear rank.
  3. [Section 4, Tables 1–2 and Figures 1, 4, 5] All reported numerical claims are based on single runs with hand-picked hyperparameters (ktrunc, η, r, p, r_sol, p_sol, τ_outer), and no standard errors or multiple random draws are provided. The comparison is also dominated by baselines from the same research group: RHOSVD-Tucker GMRES is Ref. [72] with two coauthors, and TK-SS-SD is Ref. [40] with one coauthor. Since the paper provides no convergence theory, the central claim that the proposed methods 'outperform standard low-rank Tucker solvers' rests entirely on these runs. Please add multiple trials or sensitivity studies for the main parameters (especially η and ktrunc), and include at least one independent baseline or clearly specify the settings used for the baselines.
  4. [Section 4.3, Figure 5, and Section 5] The claim that 'the low-rank Tucker constraint acts as an implicit regularizer' and that this is the first demonstration of the idea is stronger than what the experiments show. Figure 5 compares regularized versus unregularized variants of the same Tucker solvers, and Figure 4 varies the right-hand-side compression tolerance, but there is no comparison with, for example, a standard hybrid GMRES on the vectorized problem or a non-tensor regularized solver, and no formal argument (e.g., noise propagation or regularization property of the rank-truncation operator) is provided. The observed semiconvergence behavior is suggestive, but the abstract's claim of 'stable reconstructions' should either be supported by a theorem or stated more cautiously.
minor comments (6)
  1. [Section 4.2 and Table 2] The target tolerance is inconsistent: the prose states a relative residual target of 5×10^{-6}, while Table 2's caption states 5×10^{-5}. This should be reconciled, since it changes the interpretation of the reported premature convergence.
  2. [Section 3.3 and Algorithm 6] Algorithm 6 line 3 generates matrices Ψ_k ∈ R^{n_k × r}, whereas Section 3.3 defines the left sketching matrices as n_k × (r + p) and later uses ̃r = r + p in the memory comparison; the relationship between the generated matrices, the oversampling parameter p, and the sketch size ̃r should be clarified.
  3. [Section 3.3, Eq. (15)] The notation for U^{(j)}_{Ψ_k} and the barred variant used for the first r rows is nearly indistinguishable in print, and the two symbols are used interchangeably in later equations; please use clearly distinct symbols.
  4. [Section 4.3] The cross-references are incorrect: the text says 'Figure 4(a) displays the RRE over 100 iterations', but the RRE histories with and without regularization are shown in Figure 5(a)–(b), while Figure 4 concerns the effect of the RHS compression tolerance.
  5. [Section 3.4] The symbol η is overloaded: it denotes the sketched-residual correction factor in Algorithms 2 and 6, but in Eq. (24) it is also used for the exponential-sum parameter η = π/√q. Please rename one of the two.
  6. [Section 4.3] The text refers to 'tol = 0.05' when reporting the rank-(80,57,3) Tucker tensor and the RRE histories, while the algorithm parameter for right-hand-side compression is τ_outer; please align the notation and state explicitly that τ_outer = tol.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the algorithms are explicit compositions of cited building blocks, and the empirical claims rest on experiments rather than on a self-referential derivation.

full rationale

The paper makes no first-principles claim that reduces to its own inputs. RHOSVD-Tucker sGMRES and MLN-Tucker sGMRES are defined by explicit pseudo-code (Algorithms 2 and 6) that combine the sGMRES least-squares formulation (11) with the Khatri-Rao sketching and MLN subroutines from prior work; the identities used (e.g., diag(GΨ) = (Ψ1⊙...⊙Ψd)^T vec(Σ cjX(j)) in Section 3.3) are elementary algebraic equivalences, not fitted assumptions. The reported speed and memory advantages are empirical comparisons in Section 4, and the baselines (RHOSVD-Tucker GMRES from the authors' own reference [72] and TK-SS-SD from co-author Iannacito et al.) are in-group; this weakens the independence of the benchmark but is not circular reduction. The hand-tuned correction factor η in Lines 19 and 29 of Algorithms 2 and 6 is a robustness parameter, and the paper explicitly warns that 'the sketched residual may underestimate the true residual so a stopping criterion based directly on the prescribed tolerance tol may lead to premature termination,' which is exactly what Section 4.2 reports. That is an acknowledged correctness risk, not a hidden fit presented as a prediction. No uniqueness theorem or ansatz is imported from the authors' own work to force the chosen construction.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The paper contributes algorithms rather than new theory. It depends on standard randomized linear algebra assumptions (subspace embedding properties), the reliability of low-rank truncation in iterative tensor methods, and the choice of several hyperparameters. It introduces no new mathematical entities. The main burden is the unproven embedding property for structured Khatri-Rao sketches and the hand-tuning of the stopping rule.

free parameters (6)
  • ktrunc = 1 or 2
    Orthogonalization window size chosen per problem; in Section 4.2 they test ktrunc in {1,2,3} and report stagnation for ktrunc=1.
  • eta = 0.3 or 0.4
    Sketch residual correction factor; Section 3.2 notes it is introduced to avoid premature termination, and Section 4.2 shows 0.3 was insufficient for RHOSVD-Tucker sGMRES.
  • Nyström rank r = 10,25,40 (Laplacian); 60 (convection-diffusion)
    Fixed target rank for MLN; chosen by hand per experiment.
  • oversampling p = 10,25,40 (Laplacian); 180 (convection-diffusion)
    Standard oversampling parameter for Nyström; chosen per experiment.
  • memory-efficient ranks r_sol, p_sol = r_sol=p_sol=10,20,30 (Laplacian); 25 (convection-diffusion)
    Downsampled sketch ranks for reconstruction; one Laplacian run deliberately uses a too-small value to demonstrate accuracy loss.
  • compression tolerance tau_outer = 0.01,0.03,0.05
    HOSVD truncation tolerance for the right-hand side in deblurring; used to demonstrate regularization effect.
assumptions (4)
  • domain assumption Khatri-Rao structured Gaussian sketches satisfy the oblivious subspace embedding property for the relevant Krylov subspaces.
    The sGMRES theory of [58] assumes a Gaussian OSE; the paper uses structured Khatri-Rao sketches without proving the embedding property for tensor Krylov subspaces (Section 3.2).
  • domain assumption Low-rank truncation in each tensor operation introduces negligible accumulated error.
    Each RoundSum operation truncates ranks; the paper assumes the accumulated error does not break the Krylov iteration (Section 2.4.1, citing [27,71]).
  • domain assumption The Tucker format is appropriate for the problems at hand: moderate order, large mode sizes, low-rank right-hand side.
    The paper assumes B is given in Tucker format with moderate order d and that multilinear ranks remain controlled (Section 2.3).
  • domain assumption Minimizing the sketched least-squares problem with a partially orthogonal basis gives a good approximation to the true GMRES solution.
    Relies on sGMRES theory that requires orthogonality, which is only partial here (Section 3.2).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Randomized Tucker-Sketched GMRES." pith.science (2026). https://pith.science/paper/364A4KOR

@misc{pith2026260811091,
  author       = {Pith},
  title        = {Pith review of: Randomized Tucker-Sketched GMRES},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/364A4KOR}},
  note         = {Machine review of arXiv:2608.11091}
}
read the original abstract

We address the problem of solving large-scale tensor-structured linear systems in the Tucker format. In this setting, standard iterative solvers such as GMRES face a fundamental bottleneck: the multilinear ranks of the Krylov basis vectors grow with the iteration count, leading to rapidly increasing tensor operation costs and memory requirements. To overcome these challenges, we propose two randomized algorithms within the sketched GMRES framework that replace full Arnoldi orthogonalization with short recurrences. The first, RHOSVD-Tucker sGMRES, uses randomized HOSVD with per-iteration rank selection, providing robustness across a wide range of problems. The second method, MLN-Tucker sGMRES, leverages the multilinear Nystr\"om approximation with a fixed rank, enabling streaming computations; the streamability of the approximation further allows, at no additional cost, a memory-efficient reconstruction of the solution from a compact sketched representation of the Krylov basis. Both methods outperform standard low-rank Tucker solvers in symmetric and non-symmetric settings. Applied to inverse problems, the low-rank Tucker constraint acts as an implicit regularizer; combined with adaptive projected Tikhonov penalization and automatic regularization parameter selection, the methods yield stable reconstructions.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

76 extracted references · 33 canonical work pages

  1. [22]

    Epperly, Raphael A

    Chris Camaño, Ethan N. Epperly, Raphael A. Meyer, and Joel A. Tropp. Faster linear algebra algorithms with structured random matrices, 2025.arXiv:2508.21189

    arXiv 2025

  2. [72]

    Efficient sketching-based summa- tion of Tucker tensors, 2026.arXiv:2603.13532

    Rudi Smith, Mirjeta Pasha, Andrés Galindo-Olarte, Hussam Al Daas, Grey Ballard, Joseph Nakao, Jing-Mei Qiu, and William Taitano. Efficient sketching-based summa- tion of Tucker tensors, 2026.arXiv:2603.13532

    work page arXiv 2026

  3. [40]

    Subspace gradient de- scent method for linear tensor equations, 2026.arXiv:2602.21974

    Martina Iannacito, Lorenzo Piccinini, and Valeria Simoncini. Subspace gradient de- scent method for linear tensor equations, 2026.arXiv:2602.21974

    work page arXiv 2026

  4. [1]

    Asante-Mensah, Andrzej Ci- chocki, Anh Huy Phan, Tohishisa Tanaka, and Ivan Oseledets

    Salman Ahmadi-Asl, Stanislav Abukhovich, Maame G. Asante-Mensah, Andrzej Ci- chocki, Anh Huy Phan, Tohishisa Tanaka, and Ivan Oseledets. Randomized algo- rithms for computation of Tucker decomposition and higher order SVD (HOSVD). IEEE Access, 9:28684–28706, 2021.doi:10.1109/ACCESS.2021.3058103

    arXiv 2021

  5. [2]

    Taimoor Ahmed, Hassan Raja, and Waheed U. Bajwa. Tensor regression using low- rank and sparse Tucker decompositions.SIAM Journal on Mathematics of Data Sci- ence, 2(4):981–1011, 2020.doi:10.1137/19M1299335

    work page doi:10.1137/19m1299335 2020

  6. [3]

    Reid, and Arvind K

    Hussam Al Daas, Grey Ballard, Paul Cazeaux, Eric Hallman, Agnieszka Międlar, Mir- jeta Pasha, Tim W. Reid, and Arvind K. Saibaba. Randomized algorithms for round- inginthetensor-trainformat.SIAM Journal on Scientific Computing, 45(1):A74–A95, 2023.doi:10.1137/21M1451191

    doi:10.1137/21m1451191 2023

  7. [4]

    Saibaba, and Bhisham Dev Verma

    Hussam Al Daas, Grey Ballard, Laura Grigori, Mariana Martinez Aguilar, Arvind K. Saibaba, and Bhisham Dev Verma. Adaptive randomized tensor train rounding using Khatri-Rao products, 2025.arXiv:2511.03598

    work page arXiv 2025

  8. [5]

    Low-rank tensor methods for partial differential equations.Acta Numerica, 32:1–121, 2023.doi:10.1017/S0962492922000125

    Markus Bachmayr. Low-rank tensor methods for partial differential equations.Acta Numerica, 32:1–121, 2023.doi:10.1017/S0962492922000125

    doi:10.1017/s0962492922000125 2023

Show all 76 references
  1. [6]

    Randomized Gram–Schmidt process with applica- tion to GMRES.SIAM Journal on Scientific Computing, 44(3):A1450–A1474, 2022

    Oleg Balabanov and Laura Grigori. Randomized Gram–Schmidt process with applica- tion to GMRES.SIAM Journal on Scientific Computing, 44(3):A1450–A1474, 2022. doi:10.1137/20M138870X

    2022 doi

  2. [7]

    A projection method to solve linear systems in tensor format.Numerical Linear Algebra with Applications, 20(1):27–43, 2013

    Jonas Ballani and Lars Grasedyck. A projection method to solve linear systems in tensor format.Numerical Linear Algebra with Applications, 20(1):27–43, 2013. doi:10.1002/nla.1818

    2013 doi

  3. [8]

    Cambridge University Press, 2025.doi:10.1017/9781009471664

    Grey Ballard and Tamara Gibson Kolda.Tensor Decompositions for Data Science. Cambridge University Press, 2025.doi:10.1017/9781009471664

    2025 doi

  4. [9]

    Constructing structured tensor priors for Bayesian inverse problems, 2024.arXiv:2406.17597

    Kim Batselier. Constructing structured tensor priors for Bayesian inverse problems, 2024.arXiv:2406.17597

    2024 arXiv

  5. [10]

    Synthetic hollow bars 3D deconvolution dataset

    Biomedical Imaging Group, EPFL. Synthetic hollow bars 3D deconvolution dataset. URL:https://bigwww.epfl.ch/deconvolution/bars/

  6. [11]

    Brown and Richard C

    Mark A. Brown and Richard C. Semelka.MRI: Basic principles and applications. John Wiley & Sons, 2011

    2011

  7. [12]

    Randomized flexible LSQR and LSMR with applications to inverse problems, 2026.arXiv:2605.22281

    Alberto Bucci, Silvia Gazzola, and Leonardo Robol. Randomized flexible LSQR and LSMR with applications to inverse problems, 2026.arXiv:2605.22281

    2026 arXiv

  8. [13]

    Alberto Bucci and Behnam Hashemi. A sequential multilinear Nyström algorithm for streaming low-rank approximation of tensors in tucker format.Applied Mathematics Letters, 159:109271, 2025.doi:10.1016/j.aml.2024.109271

    2025

  9. [14]

    Randomized Tucker-sketched GMRES

    Alberto Bucci, Martina Iannacito, Mirjeta Pasha, and Rudi Smith. Code, data and results for numerical experiments in “Randomized Tucker-sketched GMRES” (version 1.0), August 2026.doi:10.5281/zenodo.19074856. Preprint. 2026-08-12 A. Bucci, M. Iannacito, M. Pasha, R. Smith: Rand...

    2026 doi

  10. [15]

    Randomized sketched TT- GMRES for linear systems with tensor structure.SIAM Journal on Scientific Com- puting, 47(5):A2801–A2827, 2025.doi:10.1137/24M1694999

    Alberto Bucci, Davide Palitta, and Leonardo Robol. Randomized sketched TT- GMRES for linear systems with tensor structure.SIAM Journal on Scientific Com- puting, 47(5):A2801–A2827, 2025.doi:10.1137/24M1694999

    2025 doi

  11. [16]

    Alberto Bucci and Leonardo Robol. A multilinear Nyström algorithm for low-rank approximation of tensors in Tucker format.SIAM Journal on Matrix Analysis and Applications, 45(4):1929–1953, 2024.doi:10.1137/23M1599343

    1929 doi

  12. [17]

    Randomized Algorithms for Streaming Low- Rank Approximation in Tree Tensor Network Format.Numerical Linear Algebra with Applications, 2026.doi:10.1002/nla.70107

    Alberto Bucci and Gianfranco Verzella. Randomized Algorithms for Streaming Low- Rank Approximation in Tree Tensor Network Format.Numerical Linear Algebra with Applications, 2026.doi:10.1002/nla.70107

    2026 doi

  13. [18]

    Tan Bui-Thanh, Omar Ghattas, James Martin, and Georg Stadler. A computational framework for infinite-dimensional Bayesian inverse problems Part I: The linearized case, with application to global seismic inversion.SIAM Journal on Scientific Com- puting, 35(6):A2494–A2523, 2013....

    2013 doi

  14. [19]

    Tan Bui-Thanh, Omar Ghattas, James Martin, and Georg Stadler. A computational framework for infinite-dimensional Bayesian inverse problems, Part II: Stochastic PDE-constrained inference.SIAM Journal on Scientific Computing, 35(6):A2833– A2854, 2013.doi:10.1137/130934805

    2013 doi

  15. [20]

    Krylov subspace recycling with randomized sketching for matrix functions.SIAM Journal on Matrix Analysis and Applications, 45(4):1540– 1562, 2024.doi:10.1137/23M1595904

    Liam Burke and Stefan Güttel. Krylov subspace recycling with randomized sketching for matrix functions.SIAM Journal on Matrix Analysis and Applications, 45(4):1540– 1562, 2024.doi:10.1137/23M1595904

    2024 doi

  16. [21]

    Golub, and Lothar Reichel

    Daniela Calvetti, Gene H. Golub, and Lothar Reichel. Estimation of the L-curve via Lanczos bidiagonalization.BIT Numerical Mathematics, 39(4):603–619, 1999. doi:10.1023/A:1022383005969

    1999 doi

  17. [23]

    Eckart-Young

    J. Douglas Carroll and Jih-Jie Chang. Analysis of individual differences in multi- dimensional scaling via an N-way generalization of “Eckart-Young” decomposition. Psychometrika, 35(3):283–319, 1970.doi:10.1007/BF02310791

    1970 doi

  18. [24]

    Randomized algorithms for the approximations of Tucker and the tensor train decompositions.Advances in Computational Mathematics, 45(1):395–428, 2019.doi:10.1007/s10444-018-9622-8

    Maolin Che and Yimin Wei. Randomized algorithms for the approximations of Tucker and the tensor train decompositions.Advances in Computational Mathematics, 45(1):395–428, 2019.doi:10.1007/s10444-018-9622-8

    2019 doi

  19. [25]

    Anefficientrandomizedalgorithmforcomput- ing the approximate Tucker decomposition.Journal of Scientific Computing, 88(2):32, 2021.doi:10.1007/s10915-021-01545-5

    MaolinChe, YiminWei, andHongYan. Anefficientrandomizedalgorithmforcomput- ing the approximate Tucker decomposition.Journal of Scientific Computing, 88(2):32, 2021.doi:10.1007/s10915-021-01545-5

    2021 doi

  20. [26]

    A multilinear singular value decomposition.SIAM Journal on Matrix Analysis and Applications, 21(4):1253– 1278, 2000.doi:10.1137/S0895479896305696

    Lieven De Lathauwer, Bart De Moor, and Joos Vandewalle. A multilinear singular value decomposition.SIAM Journal on Matrix Analysis and Applications, 21(4):1253– 1278, 2000.doi:10.1137/S0895479896305696

    doi

  21. [27]

    S. V. Dolgov. TT-GMRES: solution to a linear system in the structured tensor format. Russian Journal of Numerical Analysis and Mathematical Modelling, 28(2):149–172, 2013.doi:10.1515/rnam-2013-0009. Preprint. 2026-08-12 A. Bucci, M. Iannacito, M. Pasha, R. Smith: Randomized Tu...

    2013 doi

  22. [28]

    Practical sketching algorithms for low-rank Tucker approximation of large tensors.Journal of Scientific Computing, 95(2):52, 2023.doi:10.1007/s10915-023-02172-y

    Wandi Dong, Gaohang Yu, Liqun Qi, and Xiaohao Cai. Practical sketching algorithms for low-rank Tucker approximation of large tensors.Journal of Scientific Computing, 95(2):52, 2023.doi:10.1007/s10915-023-02172-y

    2023 doi

  23. [29]

    On Krylov projection methods and Tikhonov regularization.Electronic Transactions on Numerical Analy- sis, 44:83–123, 2015

    Silvia Gazzola, Paolo Novati, and Maria Rosaria Russo. On Krylov projection methods and Tikhonov regularization.Electronic Transactions on Numerical Analy- sis, 44:83–123, 2015. URL:http://emis.icm.edu.pl/journals/ETNA/vol.44.2015/ pp83-123.dir/pp83-123.pdf

    2015

  24. [30]

    Golub, Michael Heath, and Grace Wahba

    Gene H. Golub, Michael Heath, and Grace Wahba. Generalized cross-validation as a method for choosing a good ridge parameter.Technometrics, 21(2):215–223, 1979. doi:10.1080/00401706.1979.10489751

    1979

  25. [31]

    Hierarchical singular value decomposition of tensors.SIAM Jour- nal on Matrix Analysis and Applications, 31(4):2029–2054, 2010.doi:10.1137/ 090764189

    Lars Grasedyck. Hierarchical singular value decomposition of tensors.SIAM Jour- nal on Matrix Analysis and Applications, 31(4):2029–2054, 2010.doi:10.1137/ 090764189

    2010

  26. [32]

    A literature survey of low- rank tensor approximation techniques.GAMM-Mitteilungen, 36(1):53–78, 2013.doi: 10.1002/gamm.201310004

    Lars Grasedyck, Daniel Kressner, and Christine Tobler. A literature survey of low- rank tensor approximation techniques.GAMM-Mitteilungen, 36(1):53–78, 2013.doi: 10.1002/gamm.201310004

    2013 doi

  27. [33]

    Hackbusch and B

    W. Hackbusch and B. N. Khoromskij. Low-rank Kronecker-product approximation to multi-dimensional nonlocal operators. Part II. HKT representation of certain opera- tors.Computing, 76(3):203–225, January 2006.doi:10.1007/s00607-005-0145-z

    2006 doi

  28. [34]

    SIAM, Philadelphia, PA, 2010.doi:10.1137/1

    Per Christian Hansen.Discrete Inverse Problems: Insight and Algorithms, volume 7 ofFundamentals of Algorithms. SIAM, Philadelphia, PA, 2010.doi:10.1137/1. 9780898718836

    2010 doi

  29. [35]

    Per Christian Hansen, Jakob Jørgensen, and William R. B. Lionheart.Computed tomography: algorithms, insight, and just enough theory. SIAM, 2021.doi:10.1137/ 1.9781611976670

    2021

  30. [36]

    Nagy, and Dianne P

    Per Christian Hansen, James G. Nagy, and Dianne P. O’Leary.Deblurring images: matrices, spectra, and filtering. SIAM, 2006.doi:10.1137/1.9780898718874

    2006 doi

  31. [37]

    explanatory

    Richard A. Harshman. Foundations of the PARAFAC procedure: Models and conditions for an “explanatory” multi-modal factor analysis.UCLA Working Pa- pers in Phonetics, 16:1–84, 1970. University Microfilms, Ann Arbor, Michigan, No. 10,085. URL:https://three-mode.leidenuniv.nl/bib...

    1970

  32. [38]

    RTSMS: Randomized Tucker with single- mode sketching.Electronic Transactions on Numerical Analysis, 63:247–280, 2025

    Behnam Hashemi and Yuji Nakatsukasa. RTSMS: Randomized Tucker with single- mode sketching.Electronic Transactions on Numerical Analysis, 63:247–280, 2025. doi:10.1553/etna_vol63s247

    2025 doi

  33. [39]

    Higham.Functions of Matrices - Theory and Computation

    Nicholas J. Higham.Functions of Matrices - Theory and Computation. SIAM, Philadelphia, USA, 2008.doi:10.1137/1.9780898717778

    2008 doi

  34. [41]

    A practical mode-parallel implementation of the (h-)tucker decomposition via randomization, 2026.arXiv:2603.21379

    Martina Iannacito, Sascha Portaro, Davide Palitta, Claudio Arlandini, and Domitilla Brandoni. A practical mode-parallel implementation of the (h-)tucker decomposition via randomization, 2026.arXiv:2603.21379. Preprint. 2026-08-12 A. Bucci, M. Iannacito, M. Pasha, R. Smith: Ran...

    2026

  35. [42]

    Tensor-based reduced-order modeling for optimization-based inverse problems, 2026.arXiv:2607.12613

    Sahidul Islam, Andreas Mang, and Maxim Olshanskii. Tensor-based reduced-order modeling for optimization-based inverse problems, 2026.arXiv:2607.12613

    2026 arXiv

  36. [43]

    Khoromskij.Tensor numerical methods in quantum chemistry

    Venera Khoromskaia and Boris N. Khoromskij.Tensor numerical methods in quantum chemistry. Walter de Gruyter GmbH & Co KG, 2018.doi:10.1515/9783110365832

    2018 doi

  37. [44]

    Khoromskij

    Boris N. Khoromskij. Tensor numerical methods for multidimensional PDEs: theo- retical analysis and initial applications.ESAIM: Proceedings and Surveys, 48:1–28, 2015.doi:10.1051/proc/201448001

    2015

  38. [45]

    3-D PSF fitting for fluorescence microscopy: implementation and localization application.Journal of Microscopy, 249(1):13–25, 2013.doi:10.1111/j.1365-2818.2012.03675.x

    Hagai Kirshner, Franois Aguet, Daniel Sage, and Michael Unser. 3-D PSF fitting for fluorescence microscopy: implementation and localization application.Journal of Microscopy, 249(1):13–25, 2013.doi:10.1111/j.1365-2818.2012.03675.x

    2013

  39. [46]

    Kolda and Brett W

    Tamara G. Kolda and Brett W. Bader. Tensor decompositions and applications.SIAM Review, 51(3):455–500, 2009.doi:10.1137/07070111X

    2009 doi

  40. [47]

    Krylov subspace methods for linear systems with tensor product structure.SIAM Journal on Matrix Analysis and Applications, 32(4):1288–1316, 2011.doi:10.1137/100799010

    Daniel Kressner and Christine Tobler. Krylov subspace methods for linear systems with tensor product structure.SIAM Journal on Matrix Analysis and Applications, 32(4):1288–1316, 2011.doi:10.1137/100799010

    2011 doi

  41. [48]

    Streaming tensor train approximation.SIAM Journal on Scientific Computing, 45(5):A2610–A2631, 2023

    Daniel Kressner, Bart Vandereycken, and Rik Voorhaar. Streaming tensor train approximation.SIAM Journal on Scientific Computing, 45(5):A2610–A2631, 2023. doi:10.1137/22M1515045

    2023 doi

  42. [49]

    Spatiotemporal Besov priors for Bayesian inverse problems.Journal of the American Statistical Association, 2025

    Shiwei Lan, Mirjeta Pasha, Shay Li, and Weining Shen. Spatiotemporal Besov priors for Bayesian inverse problems.Journal of the American Statistical Association, 2025. doi:10.1080/01621459.2025.2560688

    2025

  43. [50]

    Priorcon- ditioned sparsity-promoting projection methods for deterministic and Bayesian linear inverse problems, 2025.arXiv:2505.01827

    Jonathan Lindbloom, Mirjeta Pasha, Jan Glaubitz, and Youssef Marzouk. Priorcon- ditioned sparsity-promoting projection methods for deterministic and Bayesian linear inverse problems, 2025.arXiv:2505.01827

    2025 arXiv

  44. [51]

    Eric F. Lock. Tensor-on-tensor regression.Journal of Computational and Graphical Statistics, 27(3):638–647, 2018.doi:10.1080/10618600.2017.1401544

    2018

  45. [52]

    Per-Gunnar Martinsson and Joel A. Tropp. Randomized numerical linear algebra: Foundations and algorithms.Acta Numerica, 29:403–572, 2020.doi:10.1017/ S0962492920000021

    2020

  46. [53]

    Shixu Meng, Daniel Appelö, and Yingda Cheng. Preconditioning low rank generalized minimal residual method (GMRES) for implicit discretizations of matrix differential equations.SIAM Journal on Matrix Analysis and Applications, 46(4):2475–2500, 2025. doi:10.1137/24M1701216

    2025 doi

  47. [54]

    Parallel randomized Tucker decompo- sition algorithms.SIAM Journal on Scientific Computing, 46(2):A1186–A1213, 2024

    Rachel Minster, Zitong Li, and Grey Ballard. Parallel randomized Tucker decompo- sition algorithms.SIAM Journal on Scientific Computing, 46(2):A1186–A1213, 2024. doi:10.1137/22M1540363

    2024 doi

  48. [55]

    Saibaba, and Misha E

    Rachel Minster, Arvind K. Saibaba, and Misha E. Kilmer. Randomized algorithms for low-rank tensor decompositions in the Tucker format.SIAM Journal on Mathematics of Data Science, 2(1):189–215, 2020.doi:10.1137/19M1261043

    2020 doi

  49. [56]

    Ronald B. Morgan. GMRES with deflated restarting.SIAM Journal on Scientific Computing, 24(1):20–37, 2002.doi:10.1137/S1064827599364659. Preprint. 2026-08-12 A. Bucci, M. Iannacito, M. Pasha, R. Smith: Randomized Tucker Sketched GMRES28

    2002 doi

  50. [57]

    Mahoney, N

    Riley Murray, James Demmel, Michael W. Mahoney, N. Benjamin Erichson, Maksim Melnichenko, Osman Asif Malik, Laura Grigori, Piotr Luszczek, Michał Dereziński, Miles E. Lopes, et al. Randomized numerical linear algebra: A perspective on the field with an eye to software, 2023.ar...

    2023 arXiv

  51. [58]

    Yuji Nakatsukasa and Joel A. Tropp. Fast and accurate randomized algorithms for linear systems and eigenvalue problems.SIAM Journal on Matrix Analysis and Ap- plications, 45(2):1183–1214, 2024.doi:10.1137/23M1565413

    2024 doi

  52. [59]

    Efficient dy- namic image reconstruction with motion estimation, 2025.arXiv:2501.12497

    Toluwani Okunola, Mirjeta Pasha, Misha Kilmer, and Melina Freitag. Efficient dy- namic image reconstruction with motion estimation, 2025.arXiv:2501.12497

    2025 arXiv

  53. [60]

    Oseledets

    Ivan V. Oseledets. Breaking the curse of dimensionality, or how to use SVD in many dimensions.SIAM Journal on Scientific Computing, 31(5):3744–3759, 2009.doi: 10.1137/090748330

    2009 doi

  54. [61]

    Oseledets

    Ivan V. Oseledets. Tensor-train decomposition.SIAM Journal on Scientific Comput- ing, 33(5):2295–2317, 2011.doi:10.1137/090752286

    2011 doi

  55. [62]

    Paige, Miroslav Rozlozník, and Zdenvek Strakos

    Christopher C. Paige, Miroslav Rozlozník, and Zdenvek Strakos. Modified Gram- Schmidt (MGS), least squares, and backward stability of MGS-GMRES.SIAM Journal on Matrix Analysis and Applications, 28(1):264–284, 2006.doi:10.1137/ 050630416

    2006

  56. [63]

    Mirjeta Pasha, Eric de Sturler, and Misha E. Kilmer. Recycling MMGKS for large- scale dynamic and streaming data, 2023.arXiv:2309.15759

    2023 arXiv

  57. [64]

    Ugwu, Jonathan Lindbloom, and Toluwani Okunola

    Mirjeta Pasha, Silvia Gazzola, Connor Sanderford, Ugochukwu O. Ugwu, Jonathan Lindbloom, and Toluwani Okunola. TRIPs-Py: Techniques for Regularization of Inverse Problems in Python.Numerical Algorithms, 2024.doi:10.1007/ s11075-024-01878-w

    2024

  58. [65]

    Saibaba, Silvia Gazzola, Malena I

    Mirjeta Pasha, Arvind K. Saibaba, Silvia Gazzola, Malena I. Español, and Eric de Sturler. A computational framework for edge-preserving regularization in dynamic inverse problems.Electronic Transactions on Numerical Analysis, 58:486–516, 2023. doi:10.1553/etna_vol58s486

    2023 doi

  59. [66]

    Pearce and Per-Gunnar Martinsson

    Katherine J. Pearce and Per-Gunnar Martinsson. Randomized algorithms for low-rank matrix and tensor decompositions, 2025.arXiv:2512.05286

    2025

  60. [67]

    A flexible inner-outer preconditioned GMRES algorithm.SIAM Journal on Scientific Computing, 14(2):461–469, 1993.doi:10.1137/0914028

    Youcef Saad. A flexible inner-outer preconditioned GMRES algorithm.SIAM Journal on Scientific Computing, 14(2):461–469, 1993.doi:10.1137/0914028

    1993 doi

  61. [68]

    Youcef Saad and Martin H. Schultz. GMRES: A generalized minimal residual al- gorithm for solving nonsymmetric linear systems.SIAM Journal on Scientific and Statistical Computing, 7(3):856–869, 1986.doi:10.1137/0907058

    1986 doi

  62. [69]

    Society for Industrial and Applied Mathematics, second edition, 2003.doi:10.1137/1.9780898718003

    Yousef Saad.Iterative Methods for Sparse Linear Systems. Society for Industrial and Applied Mathematics, second edition, 2003.doi:10.1137/1.9780898718003

    2003 doi

  63. [70]

    Kilmer, and Eric L

    Oguz Semerci, Ning Hao, Misha E. Kilmer, and Eric L. Miller. Tensor-based formu- lation and nuclear norm regularization for multienergy computed tomography.IEEE Transactions on Image Processing, 23(4):1678–1693, 2014.doi:10.1109/TIP.2014. 2305840. Preprint. 2026-08-12 A. Bucci...

    2014 doi

  64. [71]

    Valeria Simoncini and Daniel B. Szyld. Theory of inexact Krylov subspace methods and applications to scientific computing.SIAM Journal on Scientific Computing, 25(2):454–477, 2003.doi:10.1137/S1064827502406415

    2003 doi

  65. [73]

    Low-rank Tuckerapproximation of a tensor from streamingdata.SIAM Journal on Mathematics of Data Science, 2(4):1123–1150, 2020.doi:10.1137/19M1257718

    Yiming Sun, Yang Guo, Charlene Luo, Joel Tropp, and Madeleine Udell. Low-rank Tuckerapproximation of a tensor from streamingdata.SIAM Journal on Mathematics of Data Science, 2(4):1123–1150, 2020.doi:10.1137/19M1257718

    2020 doi

  66. [74]

    Ledyard R. Tucker. Some mathematical notes on three-mode factor analysis.Psy- chometrika, 31(3):279–311, 1966.doi:10.1007/BF02289464

    1966 doi

  67. [75]

    A new truncation strat- egy for the higher-order singular value decomposition.SIAM Journal on Scientific Computing, 34(2):A1027–A1052, 2012.doi:10.1137/110836067

    Nick Vannieuwenhoven, Raf Vandebril, and Karl Meerbergen. A new truncation strat- egy for the higher-order singular value decomposition.SIAM Journal on Scientific Computing, 34(2):A1027–A1052, 2012.doi:10.1137/110836067

    2012 doi

  68. [76]

    Woodruff

    David P. Woodruff. Sketching as a tool for numerical linear algebra.Foundations and Trends®in Theoretical Computer Science, 10(1–2):1–157, 2014. URL:https: //www.cs.cmu.edu/afs/cs/user/dwoodruf/www/wNow3.pdf. Preprint. 2026-08-12

    2014

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.