Pith. sign in

REVIEW 4 major objections 4 minor 1 cited by

Evidential Deep Learning for Probabilistic Modelling of Extreme Storm Events

T0 review · 4 major / 4 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 Applying evidential deep learning to storm nowcasting lets a single model pass produce calibrated predictive uncertainty at a fraction of the computational cost of ensemble or Monte Carlo dropout methods, at the price of some forecast…

desk verdict A competent application of EDL to nowcasting with an honest accuracy trade-off, but the 'enhances predictive uncertainty' claim is not yet supported by the evidence. read the letter →

arxiv 2412.14048 v1 pith:6JF4QYDD submitted 2024-12-18 cs.LG

Ayush Khot , Xihaier Luo , Ai Kagawa , Shinjae Yoo This is my paper · ORCID
classification cs.LG
keywords evidentialdeeplearninguncertaintyquantificationprecipitationnowcastingstormforecastingSEVIRdatasetEarthFormerensemblesNormal-Inverse-Gamma
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

Extreme storm forecasting needs not just a prediction but a sense of how much to trust it. Ensemble forecasts and Monte Carlo dropout provide that but require many model runs, which is expensive in real time. This paper applies evidential deep learning (EDL) to precipitation nowcasting with the EarthFormer model, producing a predictive distribution in a single pass. It reports that EDL's uncertainty is better calibrated than ten-run ensemble and MC-dropout baselines, at a fraction of the compute, though its forecast accuracy is lower. Pre-training on a plain mean-squared-error objective before adding the evidential head (P-EDL) narrows the accuracy gap while keeping most of the calibration benefit.

What carries the argument

The Normal-Inverse-Gamma (NIG) evidential head on EarthFormer—a conjugate prior distribution that lets the model output the posterior parameters in closed form. For each predicted quantity, four neurons output (γ, υ, α, β); the epistemic uncertainty is Var[µ]=β/(υ(α−1)), and this single-forward-pass uncertainty is what lets one model replace a ten-member ensemble. The loss is the negative log-likelihood of the implied Student-t distribution (Eq. 3) plus an evidence regularizer |y−γ|·(2υ+α) (Eq. 4) whose weight λ scales dynamically during training.

What would settle it

Compute EDL's epistemic uncertainty only on pixels where observed precipitation exceeds the 160 intensity threshold and compare the reliability curve to the y=x line; if it deviates more than the ensemble's curve on those extreme pixels, the paper's central calibration claim does not generalize to extremes. Alternatively, a Kolmogorov–Smirnov test of the NIG fit against SEVIR VIL pixel values would quantify the distributional mismatch.

Watch

Extended reading notes

Core claim

Replacing the deterministic output head of a spatio-temporal transformer with a Normal-Inverse-Gamma evidential head turns one forecast into a full uncertainty statement. The model outputs four quantities per pixel—(γ, υ, α, β)—so that the predicted value is γ, aleatoric uncertainty is β/(α−1), and epistemic uncertainty is β/(υ(α−1)). Training uses the negative log-likelihood of the implied Student-t distribution plus an evidence regularizer whose weight λ grows during training. On SEVIR precipitation nowcasting, the authors find that this single-pass method yields reliability curves closer to perfect calibration than deep ensembles or MC dropout, while cutting inference time and FLOPs by roughly an order of magnitude. Accuracy is worse than the baselines at high precipitation thresholds, but initializing with pretrained weights trained on MSE loss (P-EDL) improves CSI scores markedly and keeps the reliability advantage.

Load-bearing premise

The uncertainty numbers are trustworthy only if each pixel's precipitation is well described by a Gaussian with unknown mean and variance; SEVIR precipitation is mostly zeros with a heavy tail, so that distributional assumption is questionable exactly where the extremes are.

Editorial extensions

If this is right

  • If EDL's calibration holds in operation, forecasters can attach trustworthy confidence intervals to each storm nowcast without running an ensemble, making real-time warning systems cheaper.
  • P-EDL shows that a two-stage training scheme (MSE pre-training then evidential fine-tuning) recovers most of the accuracy lost by pure EDL, suggesting a practical recipe for deployment.
  • The method's single-pass cost opens the door to uncertainty quantification for very large spatio-temporal models, where running ten forward passes may be prohibitive.
  • The better-calibrated uncertainty, if verified, improves downstream risk assessment for climate and insurance applications, since decision makers can trust the model's stated confidence.

Reading between the lines

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

  • The NIG Gaussian assumption is unlikely to hold for zero-inflated, heavy-tailed precipitation fields, so the reported epistemic uncertainty may be miscalibrated precisely at the extreme intensities that matter most; a natural extension is a Gamma or hurdle-Gamma evidential head.
  • Because EDL's accuracy lags most at high CSI thresholds, an operational system might combine EDL uncertainty with a deterministic high-accuracy predictor, using the uncertainty to gate when the deterministic forecast can be trusted.
  • The dynamic λ schedule deserves scrutiny: the paper's evidence regularizer penalizes error weighted by (2υ+α), which couples the uncertainty parameters to the regression error; a stronger regularizer could trade away tail accuracy for calibration, explaining the CSI drop.
  • EDL could be used as a second stage on top of ensemble members, weighting each member by its evidential support, which the authors mention as future work but do not test.
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 / 4 minor

Summary. The paper applies evidential deep learning (EDL) to precipitation nowcasting with the EarthFormer architecture on the SEVIR dataset. The authors compare EDL (and a pretrained variant, P-EDL) against deep ensembles and MC Dropout, reporting CSI at several thresholds, MSE as a function of lead time, inference time and GFLOPS, and two uncertainty diagnostics: a normalized correlation between uncertainty and MSE and a reliability diagram. The central claim is that EDL reduces computational overhead and improves predictive uncertainty relative to ensemble-based approaches, at some cost in forecast accuracy.

Significance. If the central claim is established, the paper would provide a useful single-pass uncertainty quantification method for real-time storm nowcasting, which is a practically important application. The paper has concrete strengths: it uses a real benchmark dataset, provides a public GitHub repository, and directly measures computational cost. However, the empirical evidence does not currently establish the 'enhances predictive uncertainty' claim: the calibration comparison is only visual, the accuracy comparison appears to share identical baseline predictions, and the Gaussian likelihood underlying the EDL uncertainty is likely misspecified for precipitation data. These issues are addressable, so the work has potential, but the quantitative claims need substantially stronger support.

major comments (4)
  1. [Section 3.2, Table 1] The Ensemble and MC Dropout rows report exactly identical CSI values at all six thresholds (0.1436, 0.2613, 0.3081, 0.4225, 0.6947, 0.7666). This is implausible if the two methods are truly different: MC Dropout with 10 stochastic passes and a deep ensemble of 10 models should not produce identical binarized forecasts to four decimal places. The most likely explanation is that the same underlying deterministic predictions were used for both baselines, or that one row was copied from the other. This compromises the accuracy comparison and also undermines the later uncertainty comparison, since the baselines are not shown to be independent. Please re-run the baselines and report per-run variation or confidence intervals.
  2. [Section 3.2, Figure 3] The calibration claim rests entirely on a reliability diagram with no description of the binning scheme, the number of bins, the variable being binned, or any quantitative score such as expected calibration error, interval coverage, or CRPS. A reliability diagram can look well calibrated by eye even when important parts of the predictive distribution are wrong. Please provide quantitative calibration metrics with uncertainty estimates, and consider PIT histograms or coverage stratified by precipitation intensity threshold, since the extreme-value regime is the focus of the paper.
  3. [Section 2.2, Eqs. (2)-(5)] The EDL uncertainty model assumes that each target is Gaussian with unknown mean and variance, leading to the Normal-Inverse-Gamma formulation and the epistemic uncertainty formula Var[mu] = beta/(upsilon(alpha-1)). SEVIR VIL fields are zero-inflated (large rain-free regions) and heavy-tailed at high intensities, so this Gaussian assumption is violated exactly in the regimes the paper claims to model. The model assigns negligible probability to exact zeros and will treat extreme values as variance inflation, blurring the distinction between aleatoric and epistemic uncertainty. The authors themselves note in the Conclusion that Poisson or Gamma distributions could be adopted in EDL, which is an admission that the current likelihood choice is not settled. Please test the distributional assumption, for example with PIT diagnostics or by comparing against an EDL variant with a different likelihood, and discuss the sensitivity of the reported uncertainty calibration to this choice.
  4. [Section 3.1 and Appendix C] The empirical setup is not described in enough detail to reproduce the results. Missing information includes the training hyperparameters, number of epochs, the schedule for the regularization coefficient lambda, the number of ensemble members and their seeds, the MC Dropout rate and number of passes, the train/validation/test split, and how many repeated runs were used for any of the reported numbers. Since the paper's contribution is empirical, these details are essential for assessing the reliability of Table 1 and Figures 1-3.
minor comments (4)
  1. [Section 2.2] There is a typographical error in the sentence beginning 'Since m is composed of 4 parameters'; the word 'Since' appears to be a leftover fragment. Also, the parameter constraints (upsilon > 0, alpha > 1, beta > 0) are stated but the softplus implementation for beta is not explicitly described.
  2. [Section 3.2, Figures 1 and 3] The caption for Figure 1 reads 'average MSE for varying forecasts time leads' and should be reworded. In Figure 3, the left panel's 'normalized correlation' is not defined; please state whether this is a Pearson or Spearman correlation, normalized how, and over which pixels or events it is computed.
  3. [Appendix C.1 and Appendix A] There are several typos in the appendix, including 'synethic radar generatino', 'computional expense', 'predictoins', and 'spatiotermporal'. These should be corrected in a revision.
  4. [Section 3.2, Table 1 footnote] The footnote says P-EDL uses 'pretrained weights optimized using MSE loss', but the main text says 'initial weights pretrained without EDL'. It would be clearer to state explicitly that P-EDL is first trained with an MSE loss and then fine-tuned with the EDL loss, including the number of fine-tuning epochs.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: EDL loss is imported from external work, predictions are compared against external baselines, and no fitted quantity is renamed as a prediction.

full rationale

This paper does not attempt a first-principles derivation; it adopts the evidential deep learning objective from Amini et al. (Ref. [5]) and Sensoy et al. (Ref. [40]) in Section 2.2 (Eqs. 2-5), then evaluates the resulting single-pass uncertainty empirically against deep ensembles and MC dropout on SEVIR. The quantities compared—CSI, MSE, inference time, GFLOPS, correlation, and reliability curves—are measured outputs, not restatements of fitted parameters. No equation in the paper defines a prediction in terms of the target it is supposed to predict, and no fitted constant is later presented as a predictive result. The Normal-Inverse-Gamma Gaussian likelihood is an imported modeling assumption, not a circular reduction; whether it is appropriate for zero-inflated, heavy-tailed precipitation is a correctness and calibration question, not a circularity. The only self-citation (Ref. [30], a Bayesian deep learning weather paper) appears in the related-work survey and is not load-bearing for the central claim. The identical CSI rows for Ensemble and MC Dropout and the undescribed reliability diagram are evidence-quality concerns, not grounds for a circularity finding. The derivation chain is therefore self-contained in the relevant sense: the claim depends on external baselines and external EDL theory, not on its own outputs.

Assumptions & free parameters 2 free parameters · 4 assumptions · 0 invented entities

The paper contributes no new theory or entities. Its empirical claim depends on the EDL distributional assumption (Gaussian/NIG), the chosen regularization schedule, and a benchmark dataset; these are the main unpaid premises. The baselines' identical scores add uncertainty about whether the comparison is valid.

free parameters (2)
  • lambda (evidence regularization coefficient) = not reported
    In Eq. 5, lambda trades uncertainty inflation against model fit; the paper says it is dynamically scheduled but gives no values, and calibration depends on it.
  • number of baseline inference passes = 10
    Ensemble and MC Dropout use 10 passes; this choice directly sets the computational cost comparison.
assumptions (4)
  • domain assumption Each precipitation target is Gaussian with unknown mean and variance, with Normal-Inverse-Gamma conjugate prior.
    Section 2.2 introduces the NIG model in Eqs. 2-5; zero-inflated precipitation data likely violate this.
  • domain assumption EarthFormer is a suitable backbone and its per-pixel outputs can be reinterpreted as NIG parameters.
    The paper modifies only the output head; the backbone is taken from Ref. [17] and detailed in Appendix B.
  • domain assumption SEVIR is a representative benchmark for extreme storm nowcasting.
    The dataset is standard, but the paper does not evaluate event-level skill separately for extreme storms.
  • standard math Bayesian conjugate analysis for the NIG prior yields the Student-t negative log likelihood in Eq. 3.
    Taken from Amini et al. [5]; accepted background.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Evidential Deep Learning for Probabilistic Modelling of Extreme Storm Events." pith.science (2026). https://pith.science/paper/6JF4QYDD

@misc{pith2026241214048,
  author       = {Pith},
  title        = {Pith review of: Evidential Deep Learning for Probabilistic Modelling of Extreme Storm Events},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/6JF4QYDD}},
  note         = {Machine review of arXiv:2412.14048}
}
read the original abstract

Uncertainty quantification (UQ) methods play an important role in reducing errors in weather forecasting. Conventional approaches in UQ for weather forecasting rely on generating an ensemble of forecasts from physics-based simulations to estimate the uncertainty. However, it is computationally expensive to generate many forecasts to predict real-time extreme weather events. Evidential Deep Learning (EDL) is an uncertainty-aware deep learning approach designed to provide confidence about its predictions using only one forecast. It treats learning as an evidence acquisition process where more evidence is interpreted as increased predictive confidence. We apply EDL to storm forecasting using real-world weather datasets and compare its performance with traditional methods. Our findings indicate that EDL not only reduces computational overhead but also enhances predictive uncertainty. This method opens up novel opportunities in research areas such as climate risk assessment, where quantifying the uncertainty about future climate is crucial.

Figures

Figures reproduced from arXiv: 2412.14048 by the authors.

Figure 1
Figure 1. Plot of average MSE for varying forecasts time leads. This advantage arises because EDL conducts uncertainty analysis directly within a single model, whereas MC Dropout and Ensemble require mul￾tiple models or particles, necessitat￾ing several inferences to estimate un￾certainty. Additionally, EDL and MC Dropout have a comparable num￾ber of parameters, both significantly fewer than Ensemble, which relies on multiple… view at source ↗
Figure 2
Figure 2. Histograms of the inference time and GFLOPS. 10 20 30 40 50 60 Time Lead 0.86 0.88 0.90 0.92 0.94 0.96 0.98 1.00 Normalized Correlation EDL P-EDL Ensemble MCDO 0.0 0.2 0.4 0.6 0.8 1.0 Probability 0.0 0.2 0.4 0.6 0.8 1.0 Frequency EDL P-EDL Ensemble MCDO [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Uncertainty analysis results: The left panel displays the normalized correlation between uncertainty and MSE, while the right panel shows the reliability curves. Models closer to the dashed y = x line exhibit well-calibrated uncertainty. -bly, P-EDL maintains a higher normalized correlation over time, suggesting more stable uncertainty estimates, whereas EDL experiences a sharp decline, indicating a potential loss i… view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Example Vertically Integrated Liquid (VIL) observation sequence from the Storm EVent [PITH_FULL_IMAGE:figures/full_fig_p011_4.png]
Figure 5
Figure 5. Figure 5: Plot of error and uncertainties for EDL model [PITH_FULL_IMAGE:figures/full_fig_p012_5.png]
Figure 6
Figure 6. Figure 6: Plot of error and uncertainties for P-EDL model [PITH_FULL_IMAGE:figures/full_fig_p012_6.png]
Figure 7
Figure 7. Figure 7: Plot of error and uncertainties for EDL model [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]
Figure 8
Figure 8. Figure 8: Plot of error and uncertainties for P-EDL model [PITH_FULL_IMAGE:figures/full_fig_p013_8.png]
Figure 9
Figure 9. Figure 9: Plot of error and uncertainties for EDL model [PITH_FULL_IMAGE:figures/full_fig_p014_9.png]
Figure 10
Figure 10. Figure 10: Plot of error and uncertainties for P-EDL model [PITH_FULL_IMAGE:figures/full_fig_p014_10.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Breaking the Statistical Similarity Trap in Extreme Convection Detection

    cs.LG 2025-09 conditional novelty 5.0 of 10

    DART's dual-decoder decomposition with event-weighted training improves the critical success index for extreme convection detection from coarse atmospheric inputs, though the headline IVT ablation lacks statistical support.

Reference graph

Works this paper leans on

51 extracted references · 41 canonical work pages · cited by 1 Pith paper

  1. [1]

    Rajendra Acharya, Vladimir Makarenkov, and Saeid Nahavandi

    Moloud Abdar, Farhad Pourpanah, Sadiq Hussain, Dana Rezazadegan, Li Liu, Mohammad Ghavamzadeh, Paul Fieguth, Xiaochun Cao, Abbas Khosravi, U. Rajendra Acharya, Vladimir Makarenkov, and Saeid Nahavandi. A review of uncertainty quantification in deep learning: Techniques, applications and challenges. Information Fusion, 76:243–297, 2021

    2021

  2. [2]

    Abdullah, Masoud M

    Abdullah A. Abdullah, Masoud M. Hassan, and Yaseen T. Mustafa. A review on bayesian deep learning in healthcare: Applications and challenges. IEEE Access, 10:36538–36562, 2022

    work page 2022

  3. [3]

    Abdullah, Masoud M

    Abdullah A. Abdullah, Masoud M. Hassan, and Yaseen T. Mustafa. Leveraging bayesian deep learning and ensemble methods for uncertainty quantification in image classification: A ranking-based approach. Heliyon, 10(2):e24188, 2024

    work page 2024

  4. [4]

    Uncertainty quantification for hydrological models based on neural networks: the dropout ensemble

    Daniel Althoff, Lineu Neiva Rodrigues, and Helizani Couto Bazame. Uncertainty quantification for hydrological models based on neural networks: the dropout ensemble. Stoch. Environ. Res. Risk Assess., 35(5):1051–1067, May 2021

    work page 2021

  5. [5]

    Deep evidential regression

    Alexander Amini, Wilko Schwarting, Ava Soleimany, and Daniela Rus. Deep evidential regression. Advances in Neural Information Processing Systems, 33, 2020

    work page 2020

  6. [6]

    Adaptive precipitation nowcasting using deep learning and ensemble modeling

    Amirmasoud Amini, Mehri Dolatshahi, and Reza Kerachian. Adaptive precipitation nowcasting using deep learning and ensemble modeling. Journal of Hydrology, 612:128197, 2022

    work page 2022

  7. [7]

    Deep learning for precipitation nowcasting: A survey from the perspective of time series forecasting, 2024

    Sojung An, Tae-Jin Oh, Eunha Sohn, and Donghyun Kim. Deep learning for precipitation nowcasting: A survey from the perspective of time series forecasting, 2024

    work page 2024

  8. [8]

    Seasonal Arctic sea ice forecasting with probabilistic deep learning

    Tom R Andersson, J Scott Hosking, María Pérez-Ortiz, Brooks Paige, Andrew Elliott, Chris Russell, Stephen Law, Daniel C Jones, Jeremy Wilkinson, Tony Phillips, et al. Seasonal Arctic sea ice forecasting with probabilistic deep learning. Nature communications, 12(1):1–12, 2021

    work page 2021

Show all 51 references
  1. [9]

    Rainformer: Features extraction balanced network for radar-based precipitation nowcasting

    Cong Bai, Feng Sun, Jinglin Zhang, Yi Song, and Shengyong Chen. Rainformer: Features extraction balanced network for radar-based precipitation nowcasting. IEEE Geoscience and Remote Sensing Letters, 19:1–5, 2022

    2022

  2. [10]

    Uncertainty quantification for data-driven weather models, 2024

    Christopher Bülte, Nina Horat, Julian Quinting, and Sebastian Lerch. Uncertainty quantification for data-driven weather models, 2024

    2024

  3. [11]

    Blei and Jon D

    Alp Kucukelbir David M. Blei and Jon D. McAuliffe. Variational inference: A review for statisticians. Journal of the American Statistical Association, 112(518):859–877, 2017. 5

    2017

  4. [12]

    IFS Documentation CY47R3 - Part V Ensemble prediction system

    ECMWF. IFS Documentation CY47R3 - Part V Ensemble prediction system . Number 5. ECMWF, 09/2021 2021

    2021

  5. [13]

    Skillful twelve hour precipitation forecasts using large context neural networks

    Lasse Espeholt, Shreya Agrawal, Casper Sønderby, Manoj Kumar, Jonathan Heek, Carla Bromberg, Cenk Gazen, Jason Hickey, Aaron Bell, and Nal Kalchbrenner. Skillful twelve hour precipitation forecasts using large context neural networks. arXiv preprint arXiv:2111.07470, 2021

    2021 arXiv

  6. [14]

    Broad-unet: Multi-scale feature learning for nowcasting tasks

    Jesús García Fernández and Siamak Mehrkanoon. Broad-unet: Multi-scale feature learning for nowcasting tasks. Neural Networks, 144:419–427, 2021

    2021

  7. [15]

    Dropout as a bayesian approximation: representing model uncertainty in deep learning

    Yarin Gal and Zoubin Ghahramani. Dropout as a bayesian approximation: representing model uncertainty in deep learning. In Proceedings of the 33rd International Conference on International Conference on Machine Learning - Volume 48, ICML’16, page 1050–1059. JMLR.org, 2016

    2016

  8. [16]

    Prediff: Precipitation nowcasting with latent diffusion models

    Zhihan Gao, Xingjian Shi, Boran Han, Hao Wang, Xiaoyong Jin, Danielle C Maddix, Yi Zhu, Mu Li, and Bernie Wang. Prediff: Precipitation nowcasting with latent diffusion models. In Thirty-seventh Conference on Neural Information Processing Systems, 2023

    2023

  9. [17]

    Earthformer: Exploring space-time transformers for earth system forecasting

    Zhihan Gao, Xingjian Shi, Hao Wang, Yi Zhu, Yuyang Wang, Mu Li, and Dit-Yan Yeung. Earthformer: Exploring space-time transformers for earth system forecasting. In NeurIPS, 2022

    2022

  10. [18]

    Probabilistic forecasting

    Tilmann Gneiting and Matthias Katzfuss. Probabilistic forecasting. Annual Review of Statistics and Its Application, 1(V olume 1, 2014):125–151, 2014

    2014

  11. [19]

    Tilmann Gneiting and Adrian E. Raftery. Weather forecasting with ensemble methods. Science, 310(5746):248–249, 2005

    2005

  12. [20]

    Cascast: Skillful high-resolution precipitation nowcasting via cascaded modelling

    Junchao Gong, Lei Bai, Peng Ye, Wanghan Xu, Na Liu, Jianhua Dai, Xiaokang Yang, and Wanli Ouyang. Cascast: Skillful high-resolution precipitation nowcasting via cascaded modelling. arXiv preprint arXiv:2402.04290, 2024

    2024 arXiv

  13. [21]

    The GOES-R series: a new generation of geostationary environmental satellites

    Steven J Goodman, Timothy J Schmit, Jaime Daniels, and Robert J Redmon. The GOES-R series: a new generation of geostationary environmental satellites. Elsevier, 2019

    2019

  14. [22]

    Disentangling physical dynamics from unknown factors for unsupervised video prediction

    Vincent Le Guen and Nicolas Thome. Disentangling physical dynamics from unknown factors for unsupervised video prediction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 11474–11484, 2020

    2020

  15. [23]

    Adaptive fourier neural operators: Efficient token mixers for transformers

    John Guibas, Morteza Mardani, Zongyi Li, Andrew Tao, Anima Anandkumar, and Bryan Catanzaro. Adaptive fourier neural operators: Efficient token mixers for transformers. arXiv preprint arXiv:2111.13587, 2021

    2021 arXiv

  16. [24]

    Gustafsson, Martin Danelljan, and Thomas B

    Fredrik K. Gustafsson, Martin Danelljan, and Thomas B. Schön. Evaluating scalable bayesian deep learning methods for robust computer vision, 2020

    2020

  17. [25]

    Simple and scalable predictive uncertainty estimation using deep ensembles

    Balaji Lakshminarayanan, Alexander Pritzel, and Charles Blundell. Simple and scalable predictive uncertainty estimation using deep ensembles. InProceedings of the 31st International Conference on Neural Information Processing Systems, NIPS’17, page 6405–6416, Red Hook, NY , US...

    2017

  18. [26]

    Learning skillful medium-range global weather forecasting

    Remi Lam, Alvaro Sanchez-Gonzalez, Matthew Willson, Peter Wirnsberger, Meire Fortunato, Ferran Alet, Suman Ravuri, Timo Ewalds, Zach Eaton-Rosen, Weihua Hu, et al. Learning skillful medium-range global weather forecasting. Science, 382(6677):1416–1421, 2023

    2023

  19. [27]

    Ensemble size: How suboptimal is less than infinity? Q

    Martin Leutbecher. Ensemble size: How suboptimal is less than infinity? Q. J. R. Meteorol. Soc., 145:107–128, 9 2019. https://doi.org/10.1002/qj.3387

    2019 doi

  20. [28]

    3d high-quality magnetic resonance image restoration in clinics using deep learning

    Hao Li and Jianan Liu. 3d high-quality magnetic resonance image restoration in clinics using deep learning. ArXiv, abs/2111.14259, 2021. 6

    2021

  21. [29]

    Swin transformer: Hierarchical vision transformer using shifted windows

    Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, and Baining Guo. Swin transformer: Hierarchical vision transformer using shifted windows. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 10012–10022, 2021

    2021

  22. [30]

    A bayesian deep learning approach to near-term climate prediction

    Xihaier Luo, Balasubramanya T Nadiga, Ji Hwan Park, Yihui Ren, Wei Xu, and Shinjae Yoo. A bayesian deep learning approach to near-term climate prediction. Journal of Advances in Modeling Earth Systems, 14(10):e2022MS003058, 2022

    2022

  23. [31]

    McDermott and Christopher K

    Patrick L. McDermott and Christopher K. Wikle. Deep echo state networks with uncertainty quantification for spatio-temporal forecasting. Environmetrics, 30(3):e2553, 2019. e2553 env.2553

    2019

  24. [32]

    Ezhilarasan

    Prabha Shreeraj Nair and G. Ezhilarasan. Bayesian models for weather prediction: Using remote sensing data to improve forecast accuracy. In Om Prakash Verma, Lipo Wang, Rajesh Kumar, and Anupam Yadav, editors,Machine Intelligence for Research and Innovations, pages 327–343, Si...

    2024

  25. [33]

    Statistical field theory

    Giorgio Parisi. Statistical field theory. Addison-Wesley, 1988

    1988

  26. [34]

    FourCastNet: A global data-driven high-resolution weather model using adaptive fourier neural operators

    Jaideep Pathak, Shashank Subramanian, Peter Harrington, Sanjeev Raja, Ashesh Chattopadhyay, Morteza Mardani, Thorsten Kurth, David Hall, Zongyi Li, Kamyar Azizzadenesheli, et al. FourCastNet: A global data-driven high-resolution weather model using adaptive fourier neural oper...

    2022 arXiv

  27. [35]

    Andersson, Andrew El-Kadi, Do- minic Masters, Timo Ewalds, Jacklynn Stott, Shakir Mohamed, Peter Battaglia, Remi Lam, and Matthew Willson

    Ilan Price, Alvaro Sanchez-Gonzalez, Ferran Alet, Tom R. Andersson, Andrew El-Kadi, Do- minic Masters, Timo Ewalds, Jacklynn Stott, Shakir Mohamed, Peter Battaglia, Remi Lam, and Matthew Willson. Gencast: Diffusion-based ensemble forecasting for medium-range weather, 2024

    2024

  28. [36]

    Raftery, Tilmann Gneiting, Fadoua Balabdaoui, and Michael Polakowski

    Adrian E. Raftery, Tilmann Gneiting, Fadoua Balabdaoui, and Michael Polakowski. Us- ing bayesian model averaging to calibrate forecast ensembles. Monthly Weather Review , 133(5):1155 – 1174, 2005

    2005

  29. [37]

    Skilful pre- cipitation nowcasting using deep generative models of radar

    Suman Ravuri, Karel Lenc, Matthew Willson, Dmitry Kangin, Remi Lam, Piotr Mirowski, Megan Fitzsimons, Maria Athanassiadou, Sheleem Kashem, Sam Madge, et al. Skilful pre- cipitation nowcasting using deep generative models of radar. Nature, 597(7878):672–677, 2021

    2021

  30. [38]

    Deep learning and process understanding for data-driven earth system science

    Markus Reichstein, Gustau Camps-Valls, Bjorn Stevens, Martin Jung, Joachim Denzler, Nuno Carvalhais, et al. Deep learning and process understanding for data-driven earth system science. Nature, 566(7743):195–204, 2019

    2019

  31. [39]

    Schreck, David John Gagne II au2, Charlie Becker, William E

    John S. Schreck, David John Gagne II au2, Charlie Becker, William E. Chapman, Kim Elmore, Da Fan, Gabrielle Gantos, Eliot Kim, Dhamma Kimpara, Thomas Martin, Maria J. Molina, Vanessa M. Pryzbylo, Jacob Radford, Belen Saavedra, Justin Willson, and Christopher Wirz. Evidential d...

    2024

  32. [40]

    Evidential deep learning to quantify classification uncertainty

    Murat Sensoy, Lance Kaplan, and Melih Kandemir. Evidential deep learning to quantify classification uncertainty. In Advances in Neural Information Processing Systems, pages 3179– 3189, 2018

    2018

  33. [41]

    Convolutional LSTM network: A machine learning approach for precipitation nowcasting

    Xingjian Shi, Zhourong Chen, Hao Wang, Dit-Yan Yeung, Wai-Kin Wong, and Wang-chun Woo. Convolutional LSTM network: A machine learning approach for precipitation nowcasting. In NeurIPS, volume 28, 2015

    2015

  34. [42]

    Mc Lean Sloughter, Adrian E

    J. Mc Lean Sloughter, Adrian E. Raftery, Tilmann Gneiting, and Chris Fraley. Probabilistic quantitative precipitation forecasting using bayesian model averaging. Monthly Weather Review, 135(9):3209 – 3220, 2007

    2007

  35. [43]

    Reliable oral cancer classification framework with bayesian deep learning

    Bofan Song, Sumsum Sunny, Shaobai Li, G Keerthi, Sanjana Patrick, Nirza Mukhia, Shubha Gurudath, Subhashini Raghavan, Pramila Mendonca, Tsusennaro, Shirley T Leivon, Trupti Kolur, Vivek Shetty, Vidya Bushan R, Rohan Ramesh, Vijay Pillai, Alben Sigamani, Amritha 7 Suresh, moni ...

    2020

  36. [44]

    A bayesian approach for uncertainty quantification of extreme precipitation projections including climate model interdependency and nonstationary bias

    Maria Antonia Sunyer, Henrik Madsen, Dan Rosbjerg, and Karsten Arnbjerg-Nielsen. A bayesian approach for uncertainty quantification of extreme precipitation projections including climate model interdependency and nonstationary bias. Journal of Climate, 27(18):7113 – 7132, 2014

    2014

  37. [45]

    A survey on evidential deep learning for single-pass uncertainty estimation

    Dennis Ulmer. A survey on evidential deep learning for single-pass uncertainty estimation. ArXiv, abs/2110.03051, 2021

    2021 arXiv

  38. [46]

    SEVIR: A storm event imagery dataset for deep learning applications in radar and satellite meteorology

    Mark Veillette, Siddharth Samsi, and Chris Mattioli. SEVIR: A storm event imagery dataset for deep learning applications in radar and satellite meteorology. Advances in Neural Information Processing Systems, 33:22009–22019, 2020

    2020

  39. [47]

    Eidetic 3D LSTM: A model for video prediction and beyond

    Yunbo Wang, Lu Jiang, Ming-Hsuan Yang, Li-Jia Li, Mingsheng Long, and Li Fei-Fei. Eidetic 3D LSTM: A model for video prediction and beyond. In International conference on learning representations, 2018

    2018

  40. [48]

    PredRNN: A recurrent neural network for spatiotemporal predictive learning

    Yunbo Wang, Haixu Wu, Jianjin Zhang, Zhifeng Gao, Jianmin Wang, Philip Yu, and Mingsheng Long. PredRNN: A recurrent neural network for spatiotemporal predictive learning. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022

    2022

  41. [49]

    Diffcast: A unified framework via residual diffusion for precipitation nowcasting

    Demin Yu, Xutao Li, Yunming Ye, Baoquan Zhang, Chuyao Luo, Kuai Dai, Rui Wang, and Xunlai Chen. Diffcast: A unified framework via residual diffusion for precipitation nowcasting. In The IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024

    2024

  42. [50]

    Jordan, and Jianmin Wang

    Yuchen Zhang, Mingsheng Long, Kaiyuan Chen, Lanxiang Xing, Ronghua Jin, Michael I. Jordan, and Jianmin Wang. Skilful nowcasting of extreme precipitation with nowcastnet. Nature, 619(7970):526–532, Jul 2023

    2023

  43. [51]

    The development of the NCEP global ensemble forecast system version 12

    Xiaqiong Zhou, Yuejian Zhu, Dingchen Hou, Bing Fu, Wei Li, Hong Guan, Eric Sinsky, Walter Kolczynski, Xianwu Xue, Yan Luo, Jiayi Peng, Bo Yang, Vijay Tallapragada, and Philip Pegion. The development of the NCEP global ensemble forecast system version 12. Weather Forecast., 37(...

    2022

Pith tools

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