Pith. sign in

REVIEW 4 major objections 6 minor 87 references

Online Physics-Informed Dynamic Mode Decomposition: Theory and Applications

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read OPIDMD recasts physics-constrained DMD as convex online optimization and reports an R² of 0.991 on a noisy Lorenz forecast.

desk verdict Useful online extension of piDMD, but the global-convergence guarantee is unsupported and the Lorenz result is partly self-fulfilling. read the letter →

arxiv 2412.03609 v2 pith:V73PP6KF submitted 2024-12-04 cs.LG nlin.AO

Biqi Chen , Ying Wang This is my paper · ORCID
classification cs.LGnlin.AO
keywords dynamicmodedecompositionphysics-informedmachinelearningonlineproximalgradientdescentconvexoptimizationBayesianDMDtime-varyingdynamicalsystemsshort-termforecasting
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 introduces Online Physics-informed DMD (OPIDMD), an algorithm that solves physics-constrained Dynamic Mode Decomposition as a convex optimization problem and updates the learned linear operator one data point at a time. The authors' central claim is that this formulation converges to a unique global optimum, and that the resulting time-varying operator gives the best short-term forecasting among Exact DMD, Online DMD, and piDMD across the tested systems, including an R² of 0.991 on a noisy Lorenz system. The wider promise is a real-time solver for physics-informed DMD that scales to streaming data and resists noise.

What carries the argument

The load-bearing object is the online proximal gradient step $A_{k+1} = \operatorname{prox}_{tR}\big(A_k - t\nabla g(A_k)\big)$ for the composite objective $f(A) = \|Y_k - A_k X_k\|_F^2 + \lambda R(A_k)$, where $\operatorname{prox}_{tR}$ is the proximal mapping associated with the physical constraint. Each structural constraint receives its own proximal operator: projection onto circulant, symmetric, upper-triangular, or tridiagonal matrices; singular-value soft thresholding for the trace norm; elementwise soft thresholding for the $\ell^1$ norm; and plain gradient descent for the $\ell^2$ norm. The same convex framework is interpreted probabilistically through Bayesian MAP estimation, with the regularizer playing the role of a matrix prior. A bias-variance calculation for the $\ell^2$ case shows that stronger regularization shrinks the eigenvalues of the estimator and reduces variance at the cost of added bias.

What would settle it

Train OPIDMD with a fixed step size on a stream generated by a linear system whose true operator switches between two matrices at a known time, then compare the final iterate $A_k$ against the exact minimizer of the full-batch convex problem using all data seen so far; if they differ materially, or if the per-step objective does not settle, the global-optimality claim is falsified.

Watch

Extended reading notes

Core claim

The paper's core discovery is that the physics-informed DMD problem can be reformulated as minimizing $\|Y - AX\|_F^2 + \lambda R(A)$, with physical constraints represented as convex regularizers or projection sets, and then solved online by proximal gradient descent. Under Slater's condition the constrained and penalized forms are equivalent, and the Bayesian MAP interpretation shows that the penalty term acts as a prior. Because the data arrives sequentially, each update only needs the latest snapshot pair, and the method can accommodate implicit, circulant, symmetric, triangular, tridiagonal, low-rank, $\ell^1$, and $\ell^2$ constraints. The paper reports that OPIDMD's time-varying linear operator acts as a local linearization, which is why short-term forecasting on the chaotic Lorenz system reaches an $R^2$ of 0.991 while Exact DMD, Online DMD, and piDMD stay near 0.06.

Load-bearing premise

The claim that the method converges to a unique global optimum rests on the assumption that online proximal gradient descent with a fixed step size converges to the optimum of the entire accumulated objective even though the objective is rebuilt from a new data point at every step, and no regret or dynamic-regret bound is supplied.

Editorial extensions

If this is right

  • Physics-informed DMD can be maintained online in $O(n^2)$ per new snapshot for most constraints, instead of re-solving the full-batch problem at every data arrival.
  • The learned operator is time-varying, so each step's $A_k$ acts as a local linear approximation of a nonlinear system; this is the mechanism behind the reported Lorenz result.
  • Slater's condition plus convexity makes the constrained piDMD problem and the penalized MAP problem equivalent, so a single convex solver covers the eight physical constraints considered.
  • Physical constraints act as a bias-variance knob: they reduce forecast variance on noisy data, and overly strong constraints add bias when they do not match the true dynamics.
  • Even without explicit constraints, online gradient descent's implicit regularization gives noise robustness and simple, low-rank-like matrices.

Reading between the lines

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

  • The paper's global-optimality claim is stated for the online objective, not for the time-varying cumulative loss; a natural test is to measure dynamic regret against the best fixed matrix in hindsight, which the paper does not present.
  • The tridiagonal constraint is chosen because the Lorenz system's true DMD matrix is nearly tridiagonal; an extension would be to select the constraint adaptively via online cross-validation rather than by inspection.
  • The bias-variance analysis is carried out only for the $\ell^2$ penalty; analogous shrinkage and sparsity-bias statements for the trace norm and $\ell^1$ norm would require separate calculations.
  • Since OGD's implicit regularization is invoked as the foundation, a precise characterization of that implicit bias for online least squares would strengthen the explanation of the smooth matrices observed in the experiments.
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

4 major / 6 minor

Summary. The paper proposes OPIDMD, an online extension of physics-informed Dynamic Mode Decomposition. It casts piDMD as a convex composite optimization problem, gives a Bayesian/MAP interpretation of common regularizers, and applies online proximal gradient descent with projections or soft-thresholding for eight constraint types. Experiments on advection, Schrödinger, advection-diffusion, cylinder flow, Lorenz, and a five-degree-of-freedom mechanical system report R² values and claim superior short-term forecasting, most notably R²=0.991 on the noisy Lorenz system. The paper also provides a bias-variance analysis for the l2-regularized estimator and releases an open-source Python implementation.

Significance. If the theoretical guarantee were established, OPIDMD would be a useful general-purpose real-time solver for physics-constrained DMD. The paper has concrete strengths: explicit algorithms for each constraint, a mostly correct derivation of constrained-versus-penalized equivalence under Slater's condition, and an open-source implementation that aids reproducibility. However, the central online-convergence claim is not supported by the current theory, and the headline Lorenz result is weakened by constraint selection after inspection of the true operator and by single-run evaluations without error bars. The paper is therefore better described as a promising engineering contribution whose current claims outrun its evidence.

major comments (4)
  1. [Section 4.2, Algorithm 1, and Appendix A.2] The claim that Algorithm 1 'ensures convergence to a unique global optimum' is not established. Algorithm 1 performs an online proximal gradient step using only the gradient of the newly arrived loss g_new(A) = ||y_new - A x_new||^2, whereas the cumulative objective in Eq. (26) changes at every iteration. Standard online convex optimization provides regret bounds, not convergence to a fixed minimizer, and with a fixed step size the iterates typically track or oscillate around a moving minimizer. The assertion in Appendix A.2 that 'the inherent nature of convex problems guarantees' convergence to a unique global optimum is therefore incorrect as stated. In addition, uniqueness is not guaranteed for the l1 and trace-norm penalized problems when the quadratic data term is not strictly convex, e.g., when a zero snapshot arrives. Please replace the abstract's guarantee with a concrete regret or dynamic-regret analysis, or modify the algorithm so that it actually minimizes the cumulative objective.
  2. [Section 5.3 and Table 4] The Lorenz R²=0.991 result is not an independent prediction of the method. The tridiagonal constraint is selected after inspecting the True DMD matrix: the text states that the local dependency 'leads the True DMD matrix to exhibit a tridiagonal structure. Hence, we verify the physical validity of the model by imposing the tridiagonal matrix as a physical constraint.' Since the constraint family is chosen using the noise-free target operator, the experiment demonstrates that the chosen family contains the true structure; it does not demonstrate that OPIDMD discovers that structure from noisy data alone. Please provide an a priori physical justification for the constraint, or use a constraint-selection protocol based only on training/validation data, and evaluate on held-out systems.
  3. [Abstract and Table 4] The abstract's claim that OPIDMD 'achieves the best prediction performance in short-term forecasting' is contradicted by the paper's own results. In Table 4, for Cylinder Flow, Exact DMD achieves R²=0.938 while OPIDMD achieves 0.935, and Section 5.2 explicitly states that 'Exact DMD proved optimal as a global linear model and achieved the best predictive performance.' The claim in the abstract should be restricted to the settings in which it actually holds, or the comparison should be re-run and reported consistently.
  4. [Section 5, Tables 4 and 5] The numerical evaluation lacks uncertainty quantification. All R² values are point estimates from a single noisy realization, with no standard errors, confidence intervals, or multiple noise seeds. Several reported differences between methods are small (e.g., 0.935 versus 0.938 for Cylinder Flow), so without repeated trials the claimed advantages are not statistically substantiated. Please report mean ± standard deviation over multiple noise realizations and state the number of runs used.
minor comments (6)
  1. [Equation (24)] The indicator function I_C(x) is defined without formally introducing the set C in the surrounding text; please clarify that C denotes the constraint set in matrix space.
  2. [Algorithm 8] Algorithm 8 is titled 'Proximal Operator for unclear Norm Constraint' and repeatedly says 'unclear norm'; this should be 'nuclear norm' throughout.
  3. [Section 5, opening paragraph] The noise model is described as '25% non-stationary Gaussian white noise,' but 'non-stationary' is not defined; please specify how the noise variance or distribution changes across time and whether the same model is used for every example.
  4. [Table 4] The ODMD R² entries are astronomically negative (e.g., -9.73 × 10^196), which suggests numerical overflow or a degenerate fit; please report these values in a more informative way, such as on a clipped scale or as 'not meaningful.'
  5. [Section 5.2] The comparison for Cylinder Flow uses all eigenvalues for piDMD but only the first 50 eigenvalues for Exact DMD, OGD, and OPIDMD; this asymmetry should be justified, since it could affect the reported R² values.
  6. [Appendix D] The complexity table lists 'Streaming DMD' and 'Online DMD' but the main text does not clearly define the exact versions used; please give precise references or pseudocode for the compared online algorithms.

Circularity Check

1 steps flagged · score 6.0 of 10

The headline Lorenz R2=0.991 partly reduces to an oracle model-selection step: the tridiagonal constraint is chosen after inspecting the same system's True DMD matrix, so the reported forecast largely checks that the chosen structure matches the target rather than independently predicting it.

  1. self definitional [Section 5.3 (Lorenz System) and Table 5; True DMD defined in the Section 5 preamble]
    "True DMD refers to the model obtained using noise-free data and the Exact DMD algorithm... Due to the chaotic nature of the Lorenz system, its sensitivity to initial conditions implies that the current state is primarily influenced by recent states, reflecting a strong local dependency. This feature leads the True DMD matrix to exhibit a tridiagonal structure. Hence, we verify the physical validity of the model by imposing the tridiagonal matrix as a physical constraint."

    The tridiagonal constraint is not an independently derived physics prior; it is read off the True DMD matrix computed on noise-free data of the same Lorenz system that the model is then asked to predict. The paper imposes this oracle-selected structure in OPIDMD and reports R2=0.991, while Table 5 shows generic constraints (OGD, l1, l2, low-rank) give only about 0.235 on the same system. The headline result therefore largely verifies that the chosen constraint family contains the target operator's structure, creating a selection-feedback loop: the model constraint is defined from the target it is supposed to predict.

full rationale

Aside from the Lorenz evaluation loop, the paper's derivations are self-contained: the MAP/proximal equivalences in Section 3 and Appendix A.1 follow from standard strong duality, and the norm penalties used are nonnegative, making the constrained-versus-Lagrange equivalence exact as stated. The advection, Schrödinger, advection-diffusion, and cylinder-flow benchmarks use constraints tied to known PDE symmetries or no constraints, which is standard physics-informed modeling rather than circularity. The global-optimality claim in Appendix A.2 is a genuine gap: online proximal gradient descent on a time-varying objective is asserted to converge to a unique fixed global optimum, but no regret or dynamic-regret bound is supplied; that is an unproven assertion, not a derivation that presupposes its conclusion. The one substantial circular element is the Lorenz showcase: the tri-diagonal constraint is selected after inspecting the same system's noise-free True DMD, and the advertised R2=0.991 is therefore a verification that a tri-diagonal model class contains the target structure. Because generic constraints on the same system score about 0.235, the headline number is carried by the oracle-selected constraint. This evaluation-level circularity does not invalidate the mathematical reductions, so the score is 6 rather than higher.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The method relies on standard convex optimization and matrix-normal priors, but the performance claims depend on ad hoc choices: a per-system constraint family, unstated hyperparameters, and an unproven online convergence guarantee.

free parameters (5)
  • Regularization parameter lambda = Not reported
    Appears in l2, l1, and nuclear-norm objectives (Equations 17, 20, 7); chosen via cross-validation or by hand, values not given in the paper.
  • Step size t_k = Not reported
    Online proximal gradient step size; fixed and small or via backtracking line search; exact values not specified.
  • Constraint type per system = e.g., tridiagonal for Lorenz
    Selected per test case to maximize prediction R2 (Tables 4 and 5), making the model family a fitted choice.
  • Rank for low-rank mode selection = e.g., first 50 eigenvalues for cylinder flow
    Number of modes used in prediction is set per experiment; not justified by a principled criterion.
  • Initial matrix A0 = Not reported
    Initial point for gradient descent, presumably zero or random; not specified.
assumptions (4)
  • domain assumption Data are generated as yi = A xi + eps_i with IID Gaussian noise (Section 3.1).
    Used to justify DMD as maximum likelihood and to enable bias-variance analysis.
  • standard math Slater's condition holds for the constrained optimization problems, so constrained and penalized forms are equivalent (Appendix A.1).
    Used to equate piDMD with MAP estimation and to legitimize penalty formulations.
  • ad hoc to paper The Lorenz system's local dependency implies that the DMD matrix is tridiagonal (Section 5.3).
    This is asserted without derivation and is used to select the constraint that gives the headline R2 = 0.991.
  • ad hoc to paper Online proximal gradient descent converges to a unique global optimum in the time-varying online setting (Appendix A.2).
    No proof or regret bound is provided; standard convex convergence results apply to a fixed objective, not a changing one.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Online Physics-Informed Dynamic Mode Decomposition: Theory and Applications." pith.science (2026). https://pith.science/paper/V73PP6KF

@misc{pith2026241203609,
  author       = {Pith},
  title        = {Pith review of: Online Physics-Informed Dynamic Mode Decomposition: Theory and Applications},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/V73PP6KF}},
  note         = {Machine review of arXiv:2412.03609}
}
abstract

Dynamic Mode Decomposition (DMD) has received increasing research attention due to its capability to analyze and model complex dynamical systems. However, it faces challenges in computational efficiency, noise sensitivity, and difficulty adhering to physical laws, which negatively affect its performance. Addressing these issues, we present Online Physics-informed DMD (OPIDMD), a novel adaptation of DMD into a convex optimization framework. This approach not only ensures convergence to a unique global optimum, but also enhances the efficiency and accuracy of modeling dynamical systems in an online setting. Leveraging the Bayesian DMD framework, we propose a probabilistic interpretation of Physics-informed DMD (piDMD), examining the impact of physical constraints on the DMD linear operator. Further, we implement online proximal gradient descent and formulate specific algorithms to tackle problems with different physical constraints, enabling real-time solutions across various scenarios. Compared with existing algorithms such as Exact DMD, Online DMD, and piDMD, OPIDMD achieves the best prediction performance in short-term forecasting, e.g. an $R^2$ value of 0.991 for noisy Lorenz system. The proposed method employs a time-varying linear operator, offering a promising solution for the real-time simulation and control of complex dynamical systems.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

87 extracted references · 63 canonical work pages

  1. [1]

    Cambridge University Press, ??? (2022)

    Brunton, S.L., Kutz, J.N.: Data-driven Science and Engineering: Machine Learning, Dynamical Systems, and Control. Cambridge University Press, ??? (2022)

    2022

  2. [2]

    Structural health monitoring 19(4), 1017–1031 (2020)

    Xu, Y., Nikitas, G., Zhang, T., Han, Q., Chryssanthopoulos, M., Bhattacharya, S., Wang, Y.: Support condition monitoring of offshore wind turbines using model updating techniques. Structural health monitoring 19(4), 1017–1031 (2020)

    2020

  3. [3]

    Mechanical Systems and Signal Processing 204, 110759 (2023)

    Fu, Y., Wang, Y.: Updating numerical models towards time domain alignment of structural dynamic responses with a limited number of sensors. Mechanical Systems and Signal Processing 204, 110759 (2023)

    2023

  4. [4]

    Ocean Engineering 245, 110404 (2022)

    Abdullahi, A., Bhattacharya, S., Li, C., Xiao, Y., Wang, Y.: Long term effect of operating loads on large monopile-supported offshore wind turbines in sand. Ocean Engineering 245, 110404 (2022)

    2022

  5. [5]

    Science of the total environment 746, 141001 (2020)

    Achillopoulou, D.V., Mitoulis, S.A., Argyroudis, S.A., Wang, Y.: Monitoring of transport infras- tructure exposed to multiple hazards: A roadmap for building resilience. Science of the total environment 746, 141001 (2020)

    2020

  6. [6]

    Journal of Computing in Civil Engineering 29(2), 04014037 (2015)

    Wang, Y., Hao, H.: Damage identification scheme based on compressive sensing. Journal of Computing in Civil Engineering 29(2), 04014037 (2015)

    2015

  7. [7]

    Measurement 223, 113772 (2023)

    Zhou, J., Yu, S., Li, H., Wang, Y., Ou, J.: Automated operational modal analysis for civil engineering structures with compressed measurements. Measurement 223, 113772 (2023)

    2023

  8. [8]

    http://arxiv.org/abs/2102.12086 Accessed 2023-12-13

    Brunton, S.L., Budiˇ si´ c, M., Kaiser, E., Kutz, J.N.: Modern Koopman Theory for Dynamical Systems. http://arxiv.org/abs/2102.12086 Accessed 2023-12-13

    arXiv 2023

Show all 87 references
  1. [9]

    SIAM, ??? (2016)

    Kutz, J.N., Brunton, S.L., Brunton, B.W., Proctor, J.L.: Dynamic Mode Decomposition: Data- driven Modeling of Complex Systems. SIAM, ??? (2016)

    2016

  2. [10]

    Theoretical and computational fluid dynamics 25, 249–259 (2011) 39

    Schmid, P.J., Li, L., Juniper, M.P., Pust, O.: Applications of the dynamic mode decomposition. Theoretical and computational fluid dynamics 25, 249–259 (2011) 39

    2011

  3. [11]

    Accessed 2023-09-13

    Schmid, P.J.: Dynamic mode decomposition of numerical and experimental data 656, 5–28 https://doi.org/10.1017/S0022112010001217 . Accessed 2023-09-13

    2023 doi

  4. [12]

    Accessed 2023-11-28

    Askham, T., Kutz, J.N.: Variable projection methods for an optimized dynamic mode decom- position 17(1), 380–416 https://doi.org/10.1137/M1124176 . Accessed 2023-11-28

    2023 doi

  5. [13]

    Accessed 2023-12-12

    Tu, J.H., Rowley, C.W., Luchtenburg, D.M., Brunton, S.L., Kutz, J.N.: On dynamic mode decomposition: Theory and applications 1(2), 391–421 https://doi.org/10.3934/jcd.2014.1.391 1312.0041. Accessed 2023-12-12

    2014 arXiv

  6. [14]

    Accessed 2023-09-18

    Proctor, J.L., Brunton, S.L., Kutz, J.N.: Dynamic mode decomposition with control 15(1), 142–161 https://doi.org/10.1137/15M1013857 . Accessed 2023-09-18

    2023 doi

  7. [15]

    Accessed 2023-11-07

    Korda, M., Mezi´ c, I.: Linear predictors for nonlinear dynamical systems: Koopman operator meets model predictive control 93, 149–160 https://doi.org/10.1016/j.automatica.2018.03.046 1611.03537. Accessed 2023-11-07

    2018 arXiv

  8. [16]

    Accessed 2023-11-03

    Bai, Z., Kaiser, E., Proctor, J.L., Kutz, J.N., Brunton, S.L.: Dynamic mode decomposition for compressive system identification 58(2), 561–574 https://doi.org/10.2514/1.J057870 . Accessed 2023-11-03

    2023 doi

  9. [17]

    http://arxiv.org/abs/1404.7592 Accessed 2023-12-13

    Grosek, J., Kutz, J.N.: Dynamic Mode Decomposition for Real-Time Background/Foreground Separation in Video. http://arxiv.org/abs/1404.7592 Accessed 2023-12-13

    2023 arXiv

  10. [18]

    Journal of neuroscience methods 258, 1–15 (2016)

    Brunton, B.W., Johnson, L.A., Ojemann, J.G., Kutz, J.N.: Extracting spatial–temporal coher- ent patterns in large-scale neural recordings using dynamic mode decomposition. Journal of neuroscience methods 258, 1–15 (2016)

    2016

  11. [19]

    International health 7(2), 139–145 (2015)

    Proctor, J.L., Eckhoff, P.A.: Discovering dynamic patterns from infectious disease data using dynamic mode decomposition. International health 7(2), 139–145 (2015)

    2015

  12. [20]

    Accessed 2023-11-28

    Rowley, C.W., Dawson, S.T.M.: Model reduction for flow analysis and control 49(1), 387–417 https://doi.org/10.1146/annurev-fluid-010816-060042 . Accessed 2023-11-28

    2023 doi

  13. [21]

    1007/s00348-016-2127-7 1507.02264

    Dawson, S.T.M., Hemati, M.S., Williams, M.O., Rowley, C.W.: Characterizing and correcting for the effect of sensor noise in the dynamic mode decomposition 57(3), 42 https://doi.org/10. 1007/s00348-016-2127-7 1507.02264. Accessed 2023-12-04

    arXiv

  14. [22]

    Journal of Fluid Mechanics 917, 18 (2021)

    Pan, S., Arnold-Medabalimi, N., Duraisamy, K.: Sparsity-promoting algorithms for the discovery of informative koopman-invariant subspaces. Journal of Fluid Mechanics 917, 18 (2021)

    2021

  15. [23]

    Accessed 2023-11-05

    Baddoo, P.J., Herrmann, B., McKeon, B.J., Nathan Kutz, J., Brunton, S.L.: Physics-informed dynamic mode decomposition 479(2271), 20220576 https://doi.org/10.1098/rspa.2022.0576 . Accessed 2023-11-05

    2022

  16. [24]

    arXiv preprint arXiv:2311.15604 (2023)

    Yina, Y., Koua, C., Jiaa, S., Lu, L., Yuana, X., Luo, Y.: Pf-dmd: Physics-fusion dynamic mode decomposition for accurate and robust forecasting of dynamical systems with imperfect data and physics. arXiv preprint arXiv:2311.15604 (2023)

    2023 arXiv

  17. [25]

    arXiv preprint arXiv:2303.04323 (2023)

    Li, S., Miyazawa, Y., Yamaguchi, K., Kevrekidis, P.G., Yang, J.: Geometry-informed dynamic mode decomposition in origami dynamics. arXiv preprint arXiv:2303.04323 (2023)

    2023 arXiv

  18. [26]

    Accessed 2024-01-31

    Sharma, P., Chung, W.T., Akoush, B., Ihme, M.: A review of physics-informed machine learning in fluid mechanics 16(5), 2343 https://doi.org/10.3390/en16052343 . Accessed 2024-01-31

    2024 doi

  19. [27]

    Accessed 2023- 08-30

    Karniadakis, G.E., Kevrekidis, I.G., Lu, L., Perdikaris, P., Wang, S., Yang, L.: Physics-informed machine learning 3(6), 422–440 https://doi.org/10.1038/s42254-021-00314-5 . Accessed 2023- 08-30

    2023 doi

  20. [28]

    Accessed 2023-11-08

    Baddoo, P.J., Herrmann, B., McKeon, B.J., Brunton, S.L.: Kernel learning for robust dynamic mode decomposition: Linear and nonlinear disambiguation optimization 478(2260), 20210830 40 https://doi.org/10.1098/rspa.2021.0830 . Accessed 2023-11-08

    2021

  21. [29]

    In: Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, pp

    Takeishi, N., Kawahara, Y., Tabei, Y., Yairi, T.: Bayesian dynamic mode decomposition. In: Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, pp. 2814–2821. International Joint Conferences on Artificial Intelligence Organization. https:/...

    doi

  22. [30]

    Neural Computation 35(1), 82–103 (2023)

    Kawashima, T., Hino, H.: Gaussian process koopman mode decomposition. Neural Computation 35(1), 82–103 (2023)

    2023

  23. [31]

    Accessed 2024-01-04

    Sashidhar, D., Kutz, J.N.: Bagging, optimized dynamic mode decomposition (bop-dmd) for robust, stable forecasting with spatial and temporal uncertainty-quantification 380(2229), 20210199 https://doi.org/10.1098/rsta.2021.0199 2107.10878. Accessed 2024-01-04

    2021

  24. [32]

    Accessed 2023-12-12

    Erichson, N.B., Mathelin, L., Brunton, S.L., Kutz, J.N.: Randomized dynamic mode decompo- sition 18(4), 1867–1891 https://doi.org/10.1137/18M1215013 1702.02912. Accessed 2023-12-12

    2023 arXiv

  25. [33]

    Accessed 2023- 11-28

    Zhang, H., Rowley, C.W., Deem, E.A., Cattafesta, L.N.: Online dynamic mode decomposition for time-varying systems 18(3), 1586–1609 https://doi.org/10.1137/18M1192329 . Accessed 2023- 11-28

    2023 doi

  26. [34]

    Journal of computational dynamics 2(2), 165–191 (2016)

    Brunton, S.L., Proctor, J.L., Tu, J.H., Kutz, J.N.: Compressed sensing and dynamic mode decomposition. Journal of computational dynamics 2(2), 165–191 (2016)

    2016

  27. [35]

    AIAA Journal 58(2), 561–574 (2020)

    Bai, Z., Kaiser, E., Proctor, J.L., Kutz, J.N., Brunton, S.L.: Dynamic mode decomposition for compressive system identification. AIAA Journal 58(2), 561–574 (2020)

    2020

  28. [36]

    Journal of Real-Time Image Processing 16(5), 1479–1492 (2019)

    Erichson, N.B., Brunton, S.L., Kutz, J.N.: Compressed dynamic mode decomposition for background modeling. Journal of Real-Time Image Processing 16(5), 1479–1492 (2019)

    2019

  29. [37]

    Computer Methods in Applied Mechanics and Engineering 393, 114800 (2022)

    Zhang, X., Ji, T., Xie, F., Zheng, H., Zheng, Y.: Unsteady flow prediction from sparse measure- ments by compressed sensing reduced order modeling. Computer Methods in Applied Mechanics and Engineering 393, 114800 (2022)

    2022

  30. [38]

    Accessed 2023-12-12

    Hemati, M.S., Williams, M.O., Rowley, C.W.: Dynamic mode decomposition for large and streaming datasets 26(11), 111701 https://doi.org/10.1063/1.4901016 . Accessed 2023-12-12

    2023 doi

  31. [39]

    Accessed 2024-02-01

    Nedzhibov, G.: Extended online dmd and weighted modifications for streaming data analysis 11(6), 114 https://doi.org/10.3390/computation11060114 . Accessed 2024-02-01

    2024 doi

  32. [40]

    International Journal of Automotive Engineering 9(2), 64–71 (2018)

    Matsumoto, D., Kiewat, M., Haag, L., Indinger, T.: Online dynamic mode decomposition meth- ods for the investigation of unsteady aerodynamics of the drivaer model (first report)-validation of dmd methods for surface forces. International Journal of Automotive Engineering 9(2),...

    2018

  33. [41]

    arXiv preprint arXiv:1908.01047 (2019)

    Alfatlawi, M., Srivastava, V.: An incremental approach to online dynamic mode decom- position for time-varying systems with applications to eeg data modeling. arXiv preprint arXiv:1908.01047 (2019)

    2019 arXiv

  34. [42]

    Nedzhibov, G.: Online dynamic mode decomposition: An alternative approach for low rank datasets. Ann. Acad. Rom. Sci. Ser. Math. Appl (2023)

    2023

  35. [43]

    Accessed 2024-01-19

    Liew, J., G¨ o¸ cmen, T., Lio, W.H., Larsen, G.C.: Streaming dynamic mode decomposition for short-term forecasting in wind farms 25(4), 719–734 https://doi.org/10.1002/we.2694 . Accessed 2024-01-19

    2024 doi

  36. [44]

    Journal of Nonlinear Science 32(1), 8 (2022)

    H´ eas, P., Herzet, C.: Low-rank dynamic mode decomposition: An exact and tractable solution. Journal of Nonlinear Science 32(1), 8 (2022)

    2022

  37. [45]

    IEEE 41 transactions on information theory 56(5), 2053–2080 (2010)

    Cand` es, E.J., Tao, T.: The power of convex relaxation: Near-optimal matrix completion. IEEE 41 transactions on information theory 56(5), 2053–2080 (2010)

    2010

  38. [46]

    Journal of Computational Physics 461, 111175 (2022)

    Jiang, L., Liu, N.: Correcting noisy dynamic mode decomposition with kalman filters. Journal of Computational Physics 461, 111175 (2022)

    2022

  39. [47]

    Springer google schola 2, 5–43 (2006)

    Bishop, C.: Pattern recognition and machine learning. Springer google schola 2, 5–43 (2006)

    2006

  40. [48]

    Bertsekas, D., Tsitsiklis, J.N.: Introduction to Probability vol. 1. Athena Scientific, ??? (2008)

    2008

  41. [49]

    Chapman and Hall/CRC, ??? (2018)

    Gupta, A.K., Nagar, D.K.: Matrix Variate Distributions. Chapman and Hall/CRC, ??? (2018)

    2018

  42. [50]

    Journal of the Royal Statistical Society Series B: Statistical Methodology 58(1), 267–288 (1996)

    Tibshirani, R.: Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society Series B: Statistical Methodology 58(1), 267–288 (1996)

    1996

  43. [51]

    Neurocomputing 459, 249–289 (2021)

    Hoi, S.C., Sahoo, D., Lu, J., Zhao, P.: Online learning: A comprehensive survey. Neurocomputing 459, 249–289 (2021)

    2021

  44. [52]

    Theory and practice of online learning 2, 109–119 (2004)

    Anderson, T.: Towards a theory of online learning. Theory and practice of online learning 2, 109–119 (2004)

    2004

  45. [53]

    Fixed-point algorithms for inverse problems in science and engineering, 185–212 (2011)

    Combettes, P.L., Pesquet, J.-C.: Proximal splitting methods in signal processing. Fixed-point algorithms for inverse problems in science and engineering, 185–212 (2011)

    2011

  46. [54]

    In: Conference on Learning Theory, pp

    Woodworth, B., Gunasekar, S., Lee, J.D., Moroshko, E., Savarese, P., Golan, I., Soudry, D., Srebro, N.: Kernel and rich regimes in overparametrized models. In: Conference on Learning Theory, pp. 3635–3673 (2020). PMLR

    2020

  47. [55]

    In: Conference on Learning Theory, pp

    HaoChen, J.Z., Wei, C., Lee, J., Ma, T.: Shape matters: Understanding the implicit bias of the noise covariance. In: Conference on Learning Theory, pp. 2315–2357 (2021). PMLR

    2021

  48. [56]

    Advances in neural information processing systems 34, 5456–5468 (2021)

    Zou, D., Wu, J., Braverman, V., Gu, Q., Foster, D.P., Kakade, S.: The benefits of implicit regularization from sgd in least squares problems. Advances in neural information processing systems 34, 5456–5468 (2021)

    2021

  49. [57]

    Cambridge university press, ??? (2004)

    Boyd, S.P., Vandenberghe, L.: Convex Optimization. Cambridge university press, ??? (2004)

    2004

  50. [58]

    Nesterov, Y.: Introductory Lectures on Convex Optimization: A Basic Course vol. 87. Springer, ??? (2013)

    2013

  51. [59]

    Princeton university press, ??? (2015)

    Rockafellar, R.T.: Convex Analysis:(PMS-28). Princeton university press, ??? (2015)

    2015

  52. [60]

    In: Proceedings of COMPSTAT’2010: 19th International Conference on Computational StatisticsParis France, August 22-27, 2010 Keynote, Invited and Contributed Papers, pp

    Bottou, L.: Large-scale machine learning with stochastic gradient descent. In: Proceedings of COMPSTAT’2010: 19th International Conference on Computational StatisticsParis France, August 22-27, 2010 Keynote, Invited and Contributed Papers, pp. 177–186 (2010). Springer

    2010

  53. [61]

    Cited on 14(8), 2 (2012)

    Hinton, G., Srivastava, N., Swersky, K.: Neural networks for machine learning lecture 6a overview of mini-batch gradient descent. Cited on 14(8), 2 (2012)

    2012

  54. [62]

    Applied Mathematics and computation 219(15), 8121–8144 (2013)

    Civicioglu, P.: Backtracking search optimization algorithm for numerical optimization problems. Applied Mathematics and computation 219(15), 8121–8144 (2013)

    2013

  55. [63]

    In: Neural Networks: Tricks of the Trade: Second Edition, pp

    Bottou, L.: Stochastic gradient descent tricks. In: Neural Networks: Tricks of the Trade: Second Edition, pp. 421–436. Springer, ??? (2012)

    2012

  56. [64]

    Journal of Open Source Software 3(22), 530 (2018)

    Demo, N., Tezzele, M., Rozza, G.: Pydmd: Python dynamic mode decomposition. Journal of Open Source Software 3(22), 530 (2018)

    2018

  57. [65]

    arXiv preprint arXiv:2402.07463 (2024) 42

    Ichinaga, S.M., Andreuzzi, F., Demo, N., Tezzele, M., Lapo, K., Rozza, G., Brunton, S.L., Kutz, J.N.: Pydmd: A python package for robust dynamic mode decomposition. arXiv preprint arXiv:2402.07463 (2024) 42

    2024 arXiv

  58. [66]

    Multiscale modeling & simulation 4(4), 1168–1200 (2005)

    Combettes, P.L., Wajs, V.R.: Signal recovery by proximal forward-backward splitting. Multiscale modeling & simulation 4(4), 1168–1200 (2005)

    2005

  59. [67]

    Advances in neural information processing systems 24 (2011)

    Schmidt, M., Roux, N., Bach, F.: Convergence rates of inexact proximal-gradient methods for convex optimization. Advances in neural information processing systems 24 (2011)

    2011

  60. [68]

    In: International Conference on Machine Learning, pp

    Suzuki, T.: Dual averaging and proximal gradient descent for online alternating direction multiplier method. In: International Conference on Machine Learning, pp. 392–400 (2013). PMLR

    2013

  61. [69]

    Advances in neural information processing systems 26 (2013)

    Johnson, R., Zhang, T.: Accelerating stochastic gradient descent using predictive variance reduction. Advances in neural information processing systems 26 (2013)

    2013

  62. [70]

    Mathematical Programming 162, 83–112 (2017)

    Schmidt, M., Le Roux, N., Bach, F.: Minimizing finite sums with the stochastic average gradient. Mathematical Programming 162, 83–112 (2017)

    2017

  63. [71]

    Advances in neural information processing systems 27 (2014)

    Defazio, A., Bach, F., Lacoste-Julien, S.: Saga: A fast incremental gradient method with support for non-strongly convex composite objectives. Advances in neural information processing systems 27 (2014)

    2014

  64. [72]

    Advances in neural information processing systems 27 (2014)

    Nitanda, A.: Stochastic proximal gradient descent with acceleration techniques. Advances in neural information processing systems 27 (2014)

    2014

  65. [73]

    In: Conference On Learning Theory, pp

    Jin, C., Netrapalli, P., Jordan, M.I.: Accelerated gradient descent escapes saddle points faster than gradient descent. In: Conference On Learning Theory, pp. 1042–1085 (2018). PMLR

    2018

  66. [74]

    Neural networks 12(1), 145–151 (1999)

    Qian, N.: On the momentum term in gradient descent learning algorithms. Neural networks 12(1), 145–151 (1999)

    1999

  67. [75]

    IEEE Transactions on Parallel and Distributed Systems 31(8), 1754–1766 (2020)

    Liu, W., Chen, L., Chen, Y., Zhang, W.: Accelerating federated learning via momentum gradient descent. IEEE Transactions on Parallel and Distributed Systems 31(8), 1754–1766 (2020)

    2020

  68. [76]

    Journal of Machine Learning Research 21(219), 1–30 (2020)

    Ward, R., Wu, X., Bottou, L.: Adagrad stepsizes: Sharp convergence over nonconvex landscapes. Journal of Machine Learning Research 21(219), 1–30 (2020)

    2020

  69. [77]

    Journal of machine learning research 12(7) (2011)

    Duchi, J., Hazan, E., Singer, Y.: Adaptive subgradient methods for online learning and stochastic optimization. Journal of machine learning research 12(7) (2011)

    2011

  70. [78]

    arXiv preprint arXiv:1412.6980 (2014)

    Kingma, D.P., Ba, J.: Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980 (2014)

    2014 arXiv

  71. [79]

    In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp

    Zou, F., Shen, L., Jie, Z., Zhang, W., Liu, W.: A sufficient condition for convergences of adam and rmsprop. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 11127–11135 (2019)

    2019

  72. [80]

    The Journal of Machine Learning Research 11, 2287–2322 (2010)

    Mazumder, R., Hastie, T., Tibshirani, R.: Spectral regularization algorithms for learning large incomplete matrices. The Journal of Machine Learning Research 11, 2287–2322 (2010)

    2010

  73. [81]

    SIAM journal on imaging sciences 2(1), 183–202 (2009)

    Beck, A., Teboulle, M.: A fast iterative shrinkage-thresholding algorithm for linear inverse problems. SIAM journal on imaging sciences 2(1), 183–202 (2009)

    2009

  74. [82]

    SIAM Journal on Mathematics of Data Science 2(4), 1167–1180 (2020)

    Belkin, M., Hsu, D., Xu, J.: Two models of double descent for weak features. SIAM Journal on Mathematics of Data Science 2(4), 1167–1180 (2020)

    2020

  75. [83]

    Annals of statistics 50(2), 949 (2022)

    Hastie, T., Montanari, A., Rosset, S., Tibshirani, R.J.: Surprises in high-dimensional ridgeless least squares interpolation. Annals of statistics 50(2), 949 (2022)

    2022

  76. [84]

    Proceedings of the National Academy of Sciences 116(32), 15849–15854 (2019) 43

    Belkin, M., Hsu, D., Ma, S., Mandal, S.: Reconciling modern machine-learning practice and the classical bias–variance trade-off. Proceedings of the National Academy of Sciences 116(32), 15849–15854 (2019) 43

    2019

  77. [85]

    Advances In Neural Information Processing Systems 34, 27422–27433 (2021)

    Sekhari, A., Sridharan, K., Kale, S.: Sgd: The role of implicit regularization, batch-size and multiple-epochs. Advances In Neural Information Processing Systems 34, 27422–27433 (2021)

    2021

  78. [86]

    Pearson Education India, ??? (2007)

    Chopra, A.K.: Dynamics of Structures. Pearson Education India, ??? (2007)

    2007

  79. [87]

    Technical University of Denmark 7(15), 510 (2008) 44

    Petersen, K.B., Pedersen, M.S., et al.: The matrix cookbook. Technical University of Denmark 7(15), 510 (2008) 44

    2008

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.