Enhanced accuracy through ensembling of randomly initialized auto-regressive models for time-dependent PDEs

Indrashish Saha; Ishan Khurjekar; Lori Graham-Brady; Somdatta Goswami

REVIEW 2 major objections 5 minor 39 references

Enhanced accuracy through ensembling of randomly initialized auto-regressive models for time-dependent PDEs

T0 review · 2 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Randomly initialized autoregressive models, averaged per step, reduce long-horizon error in PDE surrogates across three systems.

desk verdict Useful empirical result on ensembling autoregressive PDE surrogates, but the diversity mechanism is asserted rather than measured and the theory does not match the closed-loop protocol. read the letter →

arxiv 2507.03863 v1 pith:VUW5PI5Q submitted 2025-07-05 cs.LG cs.AIphysics.comp-ph

Ishan Khurjekar , Indrashish Saha , Lori Graham-Brady , Somdatta Goswami This is my paper · ORCID

classification cs.LGcs.AIphysics.comp-ph

keywords time-dependentPDEssurrogatemodelautoregressivepredictionensembleerroraccumulationreductionmixtureofexpertsDiTTOshallowwaterequations

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

Machine-learning surrogates for time-dependent PDEs suffer from error accumulation: when the model feeds its own predictions back as inputs, small mistakes compound into large trajectory errors. The paper proposes to counter this without changing the architecture by training several copies of the same autoregressive model with different random weight initializations and averaging their per-step predictions during rollout. It claims this ensemble consistently reduces long-horizon error on three PDE-driven systems: stress evolution in a two-phase microstructure, Gray-Scott reaction-diffusion, and spherical shallow-water flow. Reported final-trajectory improvements are 18-28%, 5-30%, and 10-30% in relative L2 error over the range of individual models, with mean-absolute-error reductions of 14-20% over the best single model. If the claim holds, ensembling is a nearly free accuracy boost for long-horizon surrogate prediction.

What carries the argument

The load-bearing mechanism is a deep ensemble built on the DiTTO architecture, a UNet whose intermediate layers are conditioned by sinusoidal time embeddings and external-forcing embeddings, with temporal attention across the L-channel input stack. Each of the N members is trained on the same dataset with a different random weight initialization, and inference averages their next-field predictions before feeding the average back into the next autoregressive step. The theoretical anchor is the ensemble MSE decomposition: the cross-term over member error products determines whether averaging helps, with the lower bound of 1/N times the average member MSE if errors are uncorrelated, and no improvement if errors are perfectly correlated. The paper's argument is that random initialization produces sufficiently diverse local minima, so the cross-term is favorable and averaging suppresses compounding error.

What would settle it

Measure the pairwise correlation of member prediction errors along an autoregressive rollout for a more chaotic regime (for example, turbulent shallow-water flow) and check whether the ensemble's trajectory error falls below the best member's error; if it only matches the average member error, the claim fails. A separate check is to sweep the history length L from 1 to 10 and see whether the ensemble's gains survive outside the asserted L=3.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that a deep ensemble of randomly initialized autoregressive surrogates yields consistently lower accumulated error than any single member when rolled out autoregressively, across three physically distinct PDE systems. The base model is the DiTTO temporal transformer operator; each member takes the previous L=3 field snapshots plus temporal conditioning and predicts the next field. At every inference step the N member predictions are averaged, and this averaged field is fed back to all members for the next step. The ensemble's final-trajectory relative L2 errors are 0.0987 for the microstructure (individual models: 0.120-0.137), 0.085 for Gray-Scott (0.095-0.12), and 0.73 for shallow water (0.81-1.018), and its MAE beats the best member by at least 14-20% in every case. Averaged across all three datasets, the ensemble's MAE is 33% below the worst member and 15% below the best member.

Load-bearing premise

The load-bearing premise is that random initialization alone makes member errors sufficiently uncorrelated, and that this diversity persists under autoregressive feedback; the paper does not directly measure either, and it also asserts that L=3 history steps are an adequate state for all three systems.

Editorial extensions

If this is right

Surrogate accuracy for long-horizon PDE rollout improves without architectural changes, memory augmentation, or extra training objectives; the only added cost is training additional random seeds.
Because ensemble members train in parallel on separate GPUs, the wall-clock training cost need not scale linearly with ensemble size.
The benefit holds across smooth elastic-plastic evolution, pattern-forming reaction-diffusion, and rotating spherical flow, suggesting the mechanism is not tied to one dynamical regime.
The gain saturates with ensemble size: on Gray-Scott, raising the member count from 2 to 8 improves MAE by 28%, while going to 16 adds little.
At inference, the ensemble still costs one averaged forward pass per time step, so it retains the speed advantage over numerical solvers that motivates ML surrogates.

Reading between the lines

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

An untested extension the paper leaves implicit: member disagreement could serve as an in-situ uncertainty estimate, flagging when the autoregressive rollout has drifted out of the training distribution.
The error-decomposition view suggests a selection rule the paper does not explore: pruning members with highly correlated errors should improve the ensemble more than adding additional random seeds.
Because L=3 is asserted rather than swept, the method's portability to other PDEs is open; a history-length study would show whether the ensemble's advantage persists when the state representation is under-resolved.
All results are for 33-step trajectories; whether the ensemble's error suppression compounds, plateaus, or reverses over much longer rollouts is not established by the paper's evidence.

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.

Desk Editor's Note

read the letter

The paper is worth a look. It shows that averaging several randomly initialized DiTTO surrogates at each autoregressive step reduces end-of-trajectory error below every individual model on three PDE datasets (microstructure stress, Gray-Scott, shallow water). That is a clean, useful empirical result, and it is the first time I've seen deep ensembles validated specifically for rolled-out neural operator forecasts rather than static predictions. The code is available, the datasets are external (ABAQUS and the Well collection), and the reported gains are consistent: 18–28% RLE reduction on the microstructure problem, 5–30% on Gray-Scott, 10–30% on shallow water, with MAE reductions of at least 14–20% against the best single model.

Now the soft spots. The theoretical section (Eqs. 5–10) presents the standard bias-variance-covariance decomposition for a fixed input. But Algorithm 1 is closed-loop: at every step each model receives the same ensemble-averaged previous field as input. After the first step, the per-model errors are conditionally dependent through the shared state, so the independence argument used to justify the lower bound in Eq. 9 does not describe the actual rollout. The paper admits that analyzing diversity is beyond its scope, but then the abstract still credits 'diversity of model predictions' as the mechanism. That is overreach. The empirical improvement might partly come from averaging-induced smoothing of the shared trajectory rather than from genuinely diverse hypotheses. The results stand regardless, but the framing needs to be softened and the error correlation should be measured under the actual feedback protocol. Also missing: error bars or significance tests over test trajectories, and the ensemble size for the main experiments is never stated explicitly—it is only varied for Gray-Scott in Fig. 12. There are no comparisons against other stabilization techniques (e.g., FNO rollout regularization, time integration learning), so the absolute value of the gain relative to existing fixes is unclear.

The central empirical claim holds up as reported. The weakness is in the explanatory theory, not in the data or the reproducibility. This paper deserves a serious referee: it is useful, honest, and likely to improve with a revised mechanism section and a direct correlation analysis. I'd send it out.

Referee Report

2 major / 5 minor

Summary. The manuscript proposes applying deep ensembles to autoregressive neural surrogates for time-dependent PDEs. The method trains N copies of the DiTTO architecture with different random initializations and, at inference, recursively feeds the ensemble-averaged prediction of the previous L fields to all models. The authors evaluate the approach on stress evolution in loaded two-phase microstructures, Gray-Scott reaction-diffusion, and spherical shallow-water flows, reporting that the ensemble reduces final-trajectory relative L2 error and mean absolute error relative to every individual model in all three cases. A short analytical discussion of ensemble MSE decomposition and an ensemble-size study for Gray-Scott are also included.

Significance. Ensembling is a simple, architecture-agnostic intervention that, if effective, would be easy to adopt in existing autoregressive surrogates without changing the training objective. The paper's empirical setup is appealing: it uses external datasets (ABAQUS simulations and The Well collection), provides code and data links, and shows consistent directional improvements across three physically distinct systems, including a nontrivial pattern-formation problem. The main weakness is not the idea but the evidence: the key experimental configuration is incompletely specified, no uncertainty quantification is given, and the claimed diversity mechanism is not measured. With those gaps filled, the work would be a solid empirical contribution.

major comments (2)

[Sec. 2.1, Eqs. (8)-(10); Algorithm 1] The theoretical decomposition treats the input z as fixed and decomposes the ensemble MSE into individual error terms and cross terms. However, Algorithm 1 feeds the same ensemble-averaged field to every model at each rollout step, so after the first step the per-model errors are conditionally coupled through the shared state; Eq. (9) is therefore not a valid description of the closed-loop ensemble. The paper explicitly states that analyzing diversity is beyond its scope, yet the abstract attributes the error reduction to diversity of model predictions. Please add a direct measurement of error correlation under the feedback protocol (for example, pairwise correlations between per-model rollout errors, or a comparison with independently rolled-out trajectories averaged only at the end), or reframe the mechanism as a hypothesis. This is necessary to support the causal claim in the abstract.
[Secs. 4.1-4.3; Sec. 3.5] The main quantitative claims are not reproducible because the ensemble size N used for the reported RLE and MAE curves in Figs. 6, 8, and 10 is never stated; Fig. 12 varies N only for Gray-Scott, leaving the other two datasets unspecified. In addition, no error bars, confidence intervals, or repeated-seed statistics accompany the reported percentages (e.g., '18 to 28% less', '5 to 30% less', '10 to 30% less'). Please state N for every experiment and report variability, at minimum across ensemble seeds or via bootstrap over the test set, so that the reader can judge whether the ensemble advantage is within run-to-run noise.

minor comments (5)

[Sec. 3.5] The history length L=3 is set for all datasets with a recommendation of 3<=L<=10, but no experiment or analysis supports this choice. Since L defines the state representation used by the autoregressive model, the claim that 'only a few time steps' are needed would be strengthened by a sensitivity study or at least a justification based on the datasets' temporal correlations.
[Sec. 3.3 vs. Sec. 3.4] Equation (17) includes a periodic forcing F, but Sec. 3.4 states that the shallow-water temporal input contains no external forcing because variability is only from initial conditions. Please clarify whether F is a fixed known function and therefore excluded from the model input; as written, the two statements appear inconsistent.
[Fig. 9 caption] The caption refers to the 'reference FEM solution' for the shallow-water equation; this should be the reference numerical simulation, not a finite-element solution.
[Keywords] The keyword list includes 'mixture of experts', but the method neither implements nor evaluates MoE; if this is intended only as related work, the keyword list should be adjusted.
[Sec. 3.5] The statement that 100 epochs is sufficient for convergence is not supported by training curves; adding a short convergence plot or a sentence on how this was verified would help.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the ensemble error bound is standard external theory and the central claim is an empirical comparison on held-out data.

full rationale

The paper's derivation chain is not circular. The ensemble MSE decomposition (Eqs. 5-10) is the classical bias-variance-covariance identity, cited to Brown (2010), with assumptions stated explicitly (zero-bias, uncorrelated errors); the paper does not claim to derive the empirical improvement from the bound. The central claim—that averaging randomly initialized DiTTO models reduces autoregressive error—is supported by held-out test-set comparisons (Figs. 6, 8, 10) against individual models on externally sourced datasets (Gray-Scott and shallow-water from THE WELL collection, microstructure FEM data generated with ABAQUS). No parameter is fitted to the target metric and then reported as a prediction; the ensemble size is varied only as an ablation in Fig. 12. Self-citations, e.g. refs [15], [21], [25], supply architecture, baseline data, and prior deep-ensemble applications, but none is load-bearing for the central comparison: DiTTO is used as a fixed base architecture from external prior work, and the ensemble result is not deduced from any self-cited theorem. The acknowledged gap that error correlation 'is beyond the scope of this work' (Sec. 2.1) is a limitation about measuring the proposed mechanism, not a circular step; it concerns support for the mechanism, and is better classified as a correctness/robustness caveat than as circularity. The paper's claims are self-contained against external benchmarks, so the appropriate circularity score is 0.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim depends on the diversity of randomly initialized models, the Markov property of an L-step history, and the DiTTO base architecture. No physical constants are fit, and the paper introduces no new entities.

free parameters (3)

history length L = 3
Chosen uniformly across all datasets; Section 3.5 recommends 3 to 10 but gives no per-dataset validation.
ensemble size N = not stated for main results (8 implied by Fig. 12 saturation)
Figure 12 varies 2 to 16 models, but the main experiment configuration is not explicitly documented.
temporal embedding dimension = 32
Output dimension 'obtained after hyperparameter tuning' in Section 3.4; the method depends on this conditioning embedding.

assumptions (4)

domain assumption Randomly initialized neural networks trained on the same data converge to diverse local minima with partially uncorrelated errors.
Foundation of the ensemble benefit; cited from refs [9,10,13] but not verified for these datasets.
domain assumption A sliding window of L=3 past fields plus forcing is a sufficient state representation for the three PDE systems.
Used in Eq. 2 and Algorithm 1; not validated for different L in the main results.
ad hoc to paper The DiTTO architecture (ref [25]) is an adequate base surrogate for the three systems.
The paper uses DiTTO without ablating other base architectures, so the ensemble claim may not transfer to other surrogates.
standard math The ensemble MSE decomposition assumes zero-mean errors for the lower bound in Eq. 9; in practice errors are correlated and the bound is illustrative.
The decomposition into variance and covariance terms is standard; Eqs. 8 to 10 follow from ref [30].

how reviews work

0 comments

Cite this review

Pith. "Pith review of Enhanced accuracy through ensembling of randomly initialized auto-regressive models for time-dependent PDEs." pith.science (2026). https://pith.science/paper/VUW5PI5Q

@misc{pith2026250703863,
  author       = {Pith},
  title        = {Pith review of: Enhanced accuracy through ensembling of randomly initialized auto-regressive models for time-dependent PDEs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VUW5PI5Q}},
  note         = {Machine review of arXiv:2507.03863}
}

read the original abstract

Systems governed by partial differential equations (PDEs) require computationally intensive numerical solvers to predict spatiotemporal field evolution. While machine learning (ML) surrogates offer faster solutions, autoregressive inference with ML models suffer from error accumulation over successive predictions, limiting their long-term accuracy. We propose a deep ensemble framework to address this challenge, where multiple ML surrogate models with random weight initializations are trained in parallel and aggregated during inference. This approach leverages the diversity of model predictions to mitigate error propagation while retaining the autoregressive strategies ability to capture the system's time dependent relations. We validate the framework on three PDE-driven dynamical systems - stress evolution in heterogeneous microstructures, Gray-Scott reaction-diffusion, and planetary-scale shallow water system - demonstrating consistent reduction in error accumulation over time compared to individual models. Critically, the method requires only a few time steps as input, enabling full trajectory predictions with inference times significantly faster than numerical solvers. Our results highlight the robustness of ensemble methods in diverse physical systems and their potential as efficient and accurate alternatives to traditional solvers. The codes for this work are available on GitHub (https://github.com/Graham-Brady-Research-Group/AutoregressiveEnsemble_SpatioTemporal_Evolution).

Figures

Figures reproduced from arXiv: 2507.03863 by the authors.

**Figure 2.** 2-phase microstructure dataset: The top row illustrates a repre [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗

**Figure 3.** Gray-Scott reaction diffusion dataset: Species A concentration [PITH_FULL_IMAGE:figures/full_fig_p015_3.png] view at source ↗

Figures from the paper (9 more)

**Figure 4.** Figure 4: Shallow-water equation results (Shallow Water equation dataset): [PITH_FULL_IMAGE:figures/full_fig_p017_4.png]

**Figure 5.** Figure 5: 2-phase microstructure stress field prediction with the deep en [PITH_FULL_IMAGE:figures/full_fig_p020_5.png]

**Figure 6.** Figure 6: Performance metrics for the 2-phase microstructure dataset (a). [PITH_FULL_IMAGE:figures/full_fig_p021_6.png]

**Figure 7.** Figure 7: Gray-Scott reaction species concentration with the deep ensemble [PITH_FULL_IMAGE:figures/full_fig_p022_7.png]

**Figure 8.** Figure 8: Performance metrics for the Gray-Scott diffusion reaction dataset [PITH_FULL_IMAGE:figures/full_fig_p023_8.png]

**Figure 9.** Figure 9: Shallow Water equation velocity field prediction with the deep en [PITH_FULL_IMAGE:figures/full_fig_p023_9.png]

**Figure 10.** Figure 10: Performance metrics for the Shallow Water equation dataset (a). [PITH_FULL_IMAGE:figures/full_fig_p024_10.png]

**Figure 11.** Figure 11: Performance comparison of the deep ensemble approach and [PITH_FULL_IMAGE:figures/full_fig_p024_11.png]

**Figure 12.** Figure 12: Performance comparison for different numbers of base models in [PITH_FULL_IMAGE:figures/full_fig_p025_12.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 25 canonical work pages

[1]

L. Lu, P. Jin, G. Pang, Z. Zhang, G. E. Karniadakis, Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators, Nature machine intelligence 3 (2021) 218–229

2021
[2]

Z. Li, N. B. Kovachki, K. Azizzadenesheli, K. Bhattacharya, A. Stuart, A. Anandkumar, et al., Fourier Neural Operator for Parametric Par- tial Differential Equations, in: International Conference on Learning Representations, 2021

work page 2021
[3]

Goswami, A

S. Goswami, A. D. Jagtap, H. Babaee, B. T. Susi, G. E. Karniadakis, Learning stiff chemical kinetics using extended deep neural operators, Computer Methods in Applied Mechanics and Engineering 419 (2024) 116674

work page 2024
[4]

W. Diab, M. Al-Kobaisi, Temporal Neural Operator for Modeling Time-Dependent Physical Phenomena, arXiv preprint arXiv:2504.20249 (2025)

arXiv 2025
[5]

Robinson, S

H. Robinson, S. Pawar, A. Rasheed, O. San, Physics guided neural networks for modelling of non-linear dynamics, Neural Networks 154 (2022) 333–345

work page 2022
[6]

Nayak, S

D. Nayak, S. Goswami, TI-DeepONet: Learnable Time Integration for Stable Long-Term Extrapolation, arXiv preprint arXiv:2505.17341 (2025)

arXiv 2025
[7]

Micha lowska, S

K. Micha lowska, S. Goswami, G. E. Karniadakis, S. Riemer-Sørensen, Neural operator learning for long-time integration in dynamical systems 25 with recurrent neural networks, in: 2024 International Joint Conference on Neural Networks (IJCNN), IEEE, 2024, pp. 1–8

work page 2024
[8]

J. He, S. Kushwaha, J. Park, S. Koric, D. Abueidda, I. Jasiuk, Sequen- tial deep operator networks (S-deeponet) for predicting full-field solu- tions under time-dependent loads, Engineering Applications of Artificial Intelligence 127 (2024) 107258

work page 2024

Show all 39 references

[9]

Sutskever, J

I. Sutskever, J. Martens, G. Dahl, G. Hinton, On the importance of ini- tialization and momentum in deep learning, in: International conference on machine learning, PMLR, 2013, pp. 1139–1147

2013
[10]

Huang, H.-T

J. Huang, H.-T. Yau, Dynamics of deep neural networks and neural tangent hierarchy, in: International conference on machine learning, PMLR, 2020, pp. 4542–4551

2020
[11]

Breiman, Random forests, Machine learning 45 (2001) 5–32

L. Breiman, Random forests, Machine learning 45 (2001) 5–32

2001
[12]

R. A. Jacobs, M. I. Jordan, S. J. Nowlan, G. E. Hinton, Adaptive mixtures of local experts, Neural computation 3 (1991) 79–87

1991
[13]

Lakshminarayanan, A

B. Lakshminarayanan, A. Pritzel, C. Blundell, Simple and scalable predictive uncertainty estimation using deep ensembles, Advances in neural information processing systems 30 (2017)

2017
[14]

Z. Zou, Z. Wang, G. E. Karniadakis, Learning and discovering multiple solutions using physics-informed neural networks with random initial- ization and deep ensemble, arXiv preprint arXiv:2503.06320 (2025)

2025 arXiv
[15]

I. D. Khurjekar, J. B. Harley, Sim-to-real localization: Environment resilient deep ensemble learning for guided wave damage localization, The Journal of the Acoustical Society of America 151 (2022) 1325–1336

2022
[16]

Tripura, S

T. Tripura, S. Chakraborty, A foundational neural operator that con- tinuously learns without forgetting, arXiv preprint arXiv:2310.18885 (2023)

2023 arXiv
[17]

Z. Li, D. Shu, A. Barati Farimani, Scalable transformer for PDE surro- gate modeling, Advances in Neural Information Processing Systems 36 (2023) 28010–28039. 26

2023
[18]

Hemmasian, A

A. Hemmasian, A. B. Farimani, Multi-scale time-stepping of partial differential equations with transformers, Computer Methods in Applied Mechanics and Engineering 426 (2024) 116983

2024
[19]

H. Wu, H. Luo, H. Wang, J. Wang, M. Long, Transolver: A fast transformer solver for pdes on general geometries, arXiv preprint arXiv:2402.02366 (2024)

2024 arXiv
[20]

Geneva, N

N. Geneva, N. Zabaras, Transformers for modeling physical systems, Neural Networks 146 (2022) 272–289

2022
[21]

I. Saha, A. Gupta, L. Graham-Brady, Prediction of local elasto-plastic stress and strain fields in a two-phase composite microstructure using a deep convolutional neural network, Computer Methods in Applied Mechanics and Engineering 421 (2024) 116816

2024
[22]

Micha lowska, S

K. Micha lowska, S. Goswami, G. E. Karniadakis, S. Riemer-Sørensen, DON-LSTM: Multi-resolution learning with DeepONets and long short- term memory neural networks, arXiv preprint arXiv:2310.02491 (2023)

2023 arXiv
[23]

Oommen, K

V. Oommen, K. Shukla, S. Goswami, R. Dingreville, G. E. Karniadakis, Learning two-phase microstructure evolution using neural operators and autoencoder architectures, npj Computational Materials 8 (2022) 190

2022
[24]

Q. Cao, S. Goswami, T. Tripura, S. Chakraborty, G. E. Karniadakis, Deep neural operators can predict the real-time response of floating offshore structures under irregular waves, Computers & Structures 291 (2024) 107228

2024
[25]

Ovadia, V

O. Ovadia, V. Oommen, A. Kahana, A. Peyvan, E. Turkel, G. E. Kar- niadakis, Real-time inference and extrapolation via a diffusion-inspired temporal transformer operator (ditto), arXiv preprint arXiv:2307.09072 (2023)

2023 arXiv
[26]

K. He, X. Zhang, S. Ren, J. Sun, Deep residual learning for image recognition, in: Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770–778

2016
[27]

Vaswani, N

A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, L. Kaiser, I. Polosukhin, Attention is all you need, Advances in neural information processing systems 30 (2017). 27

2017
[28]

S. Fort, H. Hu, B. Lakshminarayanan, Deep ensembles: A loss landscape perspective, arXiv preprint arXiv:1912.02757 (2019)

2019 arXiv
[29]

Sharma, V

R. Sharma, V. Shankar, Ensemble and mixture-of-experts deeponets for operator learning, arXiv preprint arXiv:2405.11907 (2024)

2024 arXiv
[30]

Brown, Ensemble Learning, Springer US, Boston, MA, 2010, pp

G. Brown, Ensemble Learning, Springer US, Boston, MA, 2010, pp. 312–320. URL: https://doi.org/10.1007/978-0-387-30164-8_252 . doi:10.1007/978-0-387-30164-8_252

2010 doi
[31]

F. N. Fritsch, J. Butland, A method for constructing local monotone piecewise cubic interpolants, SIAM journal on scientific and statistical computing 5 (1984) 300–304

1984
[32]

E. J. Barbero, Finite element analysis of composite materials using Abaqus®, CRC press, 2023

2023
[33]

I. D. Khurjekar, I. Saha, A. Gupta, L. Graham-Brady, S. Goswami, Data and code associated with the publication: Full field stress response in 2-phase microstructures under dynamic loading, 2025. URL: https: //doi.org/10.7281/T1X5ZYPI. doi:10.7281/T1X5ZYPI

2025 doi
[34]

P. Gray, S. K. Scott, Autocatalytic reactions in the isothermal, continu- ous stirred tank reactor: Oscillations and instabilities in the system a+ 2b→ 3b; b→ c, Chemical Engineering Science 39 (1984) 1087–1097

1984
[35]

Ohana, M

R. Ohana, M. McCabe, L. Meyer, R. Morel, F. Agocs, M. Beneitez, M. Berger, B. Burkhart, S. Dalziel, D. Fielding, et al., The well: a large-scale collection of diverse physics simulations for machine learning, Advances in Neural Information Processing Systems 37 (2025) 44989– 45037

2025
[36]

D. L. Williamson, J. B. Drake, J. J. Hack, R. Jakob, P. N. Swarztrauber, A standard test set for numerical approximations to the shallow water equations in spherical geometry, Journal of computational physics 102 (1992) 211–224

1992
[37]

Hersbach, B

H. Hersbach, B. Bell, P. Berrisford, S. Hirahara, A. Hor´ anyi, J. Mu˜ noz- Sabater, J. Nicolas, C. Peubey, R. Radu, D. Schepers, et al., The era5 global reanalysis, Quarterly journal of the royal meteorological society 146 (2020) 1999–2049. 28

2020
[38]

McCabe, P

M. McCabe, P. Harrington, S. Subramanian, J. Brown, Towards stabil- ity of autoregressive neural operators, arXiv preprint arXiv:2306.10619 (2023)

2023 arXiv
[39]

S. Wang, X. Yu, P. Perdikaris, When and why pinns fail to train: A neural tangent kernel perspective, Journal of Computational Physics 449 (2022) 110768. 29

2022

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.

[1] [1]

L. Lu, P. Jin, G. Pang, Z. Zhang, G. E. Karniadakis, Learning nonlinear operators via DeepONet based on the universal approximation theorem of operators, Nature machine intelligence 3 (2021) 218–229

2021

[2] [2]

Z. Li, N. B. Kovachki, K. Azizzadenesheli, K. Bhattacharya, A. Stuart, A. Anandkumar, et al., Fourier Neural Operator for Parametric Par- tial Differential Equations, in: International Conference on Learning Representations, 2021

work page 2021

[3] [3]

Goswami, A

S. Goswami, A. D. Jagtap, H. Babaee, B. T. Susi, G. E. Karniadakis, Learning stiff chemical kinetics using extended deep neural operators, Computer Methods in Applied Mechanics and Engineering 419 (2024) 116674

work page 2024

[4] [4]

W. Diab, M. Al-Kobaisi, Temporal Neural Operator for Modeling Time-Dependent Physical Phenomena, arXiv preprint arXiv:2504.20249 (2025)

arXiv 2025

[5] [5]

Robinson, S

H. Robinson, S. Pawar, A. Rasheed, O. San, Physics guided neural networks for modelling of non-linear dynamics, Neural Networks 154 (2022) 333–345

work page 2022

[6] [6]

Nayak, S

D. Nayak, S. Goswami, TI-DeepONet: Learnable Time Integration for Stable Long-Term Extrapolation, arXiv preprint arXiv:2505.17341 (2025)

arXiv 2025

[7] [7]

Micha lowska, S

K. Micha lowska, S. Goswami, G. E. Karniadakis, S. Riemer-Sørensen, Neural operator learning for long-time integration in dynamical systems 25 with recurrent neural networks, in: 2024 International Joint Conference on Neural Networks (IJCNN), IEEE, 2024, pp. 1–8

work page 2024

[8] [8]

J. He, S. Kushwaha, J. Park, S. Koric, D. Abueidda, I. Jasiuk, Sequen- tial deep operator networks (S-deeponet) for predicting full-field solu- tions under time-dependent loads, Engineering Applications of Artificial Intelligence 127 (2024) 107258

work page 2024