Pith. sign in

REVIEW 4 major objections 4 minor 50 references

Localized Weather Prediction Using Kolmogorov-Arnold Network-Based Models and Deep RNNs

T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read A default Kolmogorov-Arnold Network predicts next-day temperature in two tropical cities with $R^2$ above 0.998, far ahead of deep recurrent baselines.

desk verdict KAN temperature numbers are internally inconsistent; as written the main result is unsupported, though the benchmark setup is decent. read the letter →

arxiv 2505.22686 v1 pith:7EG667DM submitted 2025-05-27 cs.LG physics.ao-ph

Ange-Clement Akazan , Verlon Roel Mbingui , Gnankan Landry Regis N'guessan , Issa Karambal This is my paper
classification cs.LGphysics.ao-ph
keywords Kolmogorov-ArnoldNetworksweatherforecastingtropicalAfricarecurrentneuralLSTMTKANtemperaturepredictionprecipitation
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

The paper sets out to determine whether spline-based Kolmogorov-Arnold networks can outperform deep recurrent networks for next-day local weather prediction in tropical Africa. Using 14 years of daily station data from Abidjan and Kigali, it benchmarks LSTM, GRU, BiLSTM, BiGRU, an ensemble, KAN, and TKAN (including two new activation variants) on temperature, precipitation, and surface pressure. Its central finding is that the default KAN forecasts 2-meter temperature almost exactly, with $R^2$ above 0.998 and MSE below $0.0014\,^\circ\mathrm{C}^2$ in both cities, while recurrent baselines sit near $R^2 = 0.83$--$0.86$. It also reports that TKAN variants with GeLU and MiSH activations give the best precipitation errors in low-rainfall conditions, and that classical RNNs remain best for pressure. If these results hold out of sample, they point to a lightweight, data-efficient alternative to numerical weather prediction for localized forecasting.

What carries the argument

The load-bearing object is the Kolmogorov-Arnold Network, which replaces each fixed neuron activation with a learnable univariate B-spline on the edge, so a layer is a composition of spline functions rather than a matrix of weights followed by a fixed nonlinearity. The paper uses the layer form $\Phi_l = W_b\, b(x) + W_s\, \mathrm{Spline}(x)$ with the SiLU base activation, and for the temporal variant TKAN it couples these KAN sub-layers with LSTM-style forget, input, and output gates to carry memory across time steps. That machinery is what the paper credits for smooth, local, data-efficient approximations; the custom variants swap SiLU for GeLU or MiSH in the base function, and the comparison against deep RNNs is what isolates the contribution of the spline representation.

What would settle it

Re-run the KAN temperature experiment with the final 20% of the daily series held out as the test set and report MSE and $R^2$; if $R^2$ falls from about 0.999 to the 0.75--0.86 range seen for the recurrent baselines, the claimed order-of-magnitude superiority is an artifact of the split.

Watch

Extended reading notes

Core claim

On its own terms, the paper's central discovery is that a default-configured Kolmogorov-Arnold Network predicts daily 2-meter temperature from a 14-day window of ten weather variables with near-perfect accuracy: $\mathrm{MSE}=0.0014\,^\circ\mathrm{C}^2$ and $R^2=0.9986$ in Abidjan, and $\mathrm{MSE}=0.0003\,^\circ\mathrm{C}^2$ and $R^2=0.9998$ in Kigali, outperforming every recurrent baseline by roughly three orders of magnitude in mean squared error. For precipitation, the temporal KANs---especially the new GeLU and MiSH variants---produce the lowest absolute errors in Kigali, a low-rainfall regime, while KAN itself gives the highest $R^2$ on precipitation in both cities. For surface pressure, the deep RNNs (LSTM, BiLSTM, GRU, BiGRU, ensemble) remain superior, with $R^2 \approx 0.83$--$0.86$, while KAN drops to about $0.50$. The paper interprets this as evidence that spline-based architectures are data-efficient for smooth targets like temperature but need temporal memory mechanisms to compete on noisier variables.

Load-bearing premise

The benchmark's train/validation/test split is described only as 72/8/20 and not as chronological, so the near-perfect KAN temperature scores could be inflated if randomly chosen test days have near-duplicate training days.

Editorial extensions

If this is right

  • A default KAN, with no hyperparameter tuning, can serve as a lightweight next-day temperature forecaster for tropical cities using only daily station observations and a 14-day history.
  • For precipitation in low-rainfall regimes, TKAN variants using GeLU or MiSH activations are the preferred spline-based option over the standard SiLU TKAN.
  • Surface pressure forecasting should continue to rely on recurrent architectures; replacing them with plain KAN would degrade $R^2$ from about 0.85 to about 0.50.
  • The same KAN/TKAN setup transfers across two contrasting climates (coastal Abidjan and highland Kigali) without re-architecting, supporting the idea of location-agnostic, data-driven forecasting.

Reading between the lines

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

  • A direct next step is to rerun the temperature benchmark using only the final contiguous 20% of the daily series as the test set and report the KAN-versus-RNN gap in $R^2$; this would remove any ambiguity about out-of-sample evaluation.
  • The same protocol could be extended to 3- and 7-day horizons; if KAN's edge is purely one-step persistence, its advantage should shrink rapidly as the horizon grows.
  • For precipitation, reporting categorical skill (e.g., probability of detecting a rain day) alongside continuous errors would connect these benchmarks to early-warning use, since the reported MAPE values are in the hundreds of percent.
  • The pressure results suggest a hybrid architecture---RNN gates for memory plus spline layers for local approximation---might combine the strengths of both families; the paper's data do not test this directly.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 4 minor

Summary. The paper benchmarks LSTM, GRU, BiLSTM, BiGRU, an ensemble, KAN, TKAN, and two TKAN variants with GeLU and MiSH activations for one-day-ahead forecasting of temperature, precipitation, and surface pressure at two tropical African stations (Abidjan and Kigali), using 14-day windows of ten meteorological variables from NASA POWER data (2010-2024). The central claimed result is that a default KAN achieves near-perfect temperature forecasts (R² = 0.9986 in Abidjan and 0.9998 in Kigali, MSE below 0.0014 °C²), three orders of magnitude better than all recurrent baselines, while TKAN variants improve precipitation forecasts in low-rainfall regimes and classical RNNs remain competitive for pressure. The paper also introduces and tests two TKAN activation variants.

Significance. If the results were reliable, the paper would provide a useful early benchmark of KAN-family models for localized weather prediction in under-studied tropical African cities, with a concrete claim about data efficiency and a falsifiable performance comparison. The authors provide a public code link, which is a strength for reproducibility, and they compare a broad set of architectures on three meteorological variables. However, the headline temperature result is internally inconsistent with the other metrics reported in the same table, and the evaluation protocol is not demonstrably out-of-sample; both issues directly affect the central claim. As reported, the numbers cannot be accepted as evidence for KAN superiority in temperature forecasting.

major comments (4)
  1. [Section 4, Table 2] The temperature metrics for Abidjan are internally inconsistent under the standard definition R² = 1 - MSE/Var(test). The KAN row (MSE = 0.0014, R² = 0.9986) implies a test-set variance of 1.0 °C², whereas the LSTM row (MSE = 7.2616, R² = 0.8349) implies a test-set variance of 43.97 °C² on the same test set. The same discrepancy appears for Kigali, and the TKAN rows also imply a variance near 1 °C². Since a single test split cannot have two different variances, the reported MSE/R² pairs for KAN and TKAN on the one hand and the recurrent models on the other cannot all be computed on the same target scale. This suggests a mixing of scaled and original units or an incorrect R² computation, and it invalidates the abstract's central claim that KAN achieves R² ≈ 0.999 with MSE < 0.0014 °C².
  2. [Section 3.1.2] The data-split description states only that the dataset was split into 72% training, 8% validation, and 20% testing; it does not state that the split is chronological. Daily meteorological series are strongly autocorrelated, so a random split places near-identical consecutive days in both training and test sets, making the reported near-perfect temperature R² values an artifact of persistence rather than evidence of genuine forecast skill. The authors must specify whether the split respects time order and, if not, rerun all experiments with a strictly chronological split (e.g., training on earlier years and testing on later years) before any comparison between models is meaningful.
  3. [Section 4, paragraph 1] The manuscript reports that KAN was used with 'default parameters' but does not report the number of KAN layers, hidden widths, grid size, spline order, or the number of training epochs. The same applies to the TKAN variants beyond the number of sub-layers. Given that the central claim is a three-orders-of-magnitude improvement over recurrent baselines, the absence of these architectural and training details makes the result non-reproducible even with the provided code link.
  4. [Section 4, Table 1] The Kigali precipitation row for TKAN (5 Sub-layers) reports MSE = 23.5482 and R² = 0.1568, which implies a test variance of 27.93 mm², while other models in the same table imply a variance around 37.5 mm² (e.g., LSTM: 29.4953 / (1 - 0.2145) ≈ 37.55). This is another indication of inconsistent metric computation, although it is less central than the temperature discrepancy.
minor comments (4)
  1. [Section 4, paragraph 2] The phrase 'the most best performance' should be corrected to 'the best performance.'
  2. [Section 4, Table 2 and Figures 4-5] The figures labeled 'predicted vs. actual temperature' appear with small or illegible axis labels and no quantitative scale; please enlarge the axes and add units so the reader can visually verify the claimed near-perfect fit.
  3. [References] Several references are incomplete, for example [1] lacks a publication venue and year, and [8] appears to have a formatting error ('2002 2002'); please unify the bibliography style.
  4. [Section 3.1.2] The description of the scaling choices is confusing: it says MinMaxScaler [0,1] was used for T2M and PS with recurrent models, and [-1,1] for PREC, but 'for the other models, all variables were scaled to [0,1]'. This makes it unclear which models and which targets are evaluated in original units; please clarify the exact scaling and inverse-scaling procedure for each metric reported.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper is an empirical benchmark with no derivation chain whose outputs reduce to inputs by construction.

full rationale

This manuscript reports an empirical comparison of KAN/TKAN variants against RNN baselines on three meteorological targets. It does not derive a predictor from a fitted quantity, invoke a uniqueness theorem, or rename a known result as a new derivation. The central protocol is a standard supervised learning setup described in Section 3.1.2: a 72/8/20 split, MinMaxScaler for preprocessing, and a 14-day sliding window to predict the next day. Evaluation metrics in Tables 1-3 are computed on held-out test partitions, and the paper does not claim that any reported quantity is both fitted and predicted. The customized TKAN variants merely replace the SiLU activation with GeLU or MiSH (Section 3.2.6), which is a straightforward architectural modification, not a circular step. Citations such as [17] for TKAN and [34] for KAN are external references, not self-citations of the present authors, and no load-bearing claim relies on an author-imported uniqueness theorem. Possible concerns about the temperature results, such as the internal inconsistency of the reported MSE/R2 pairs and the strong correlation of T2MDEW/T2MWET inputs with the T2M target, are validity, leakage, and correctness issues rather than circularity: they do not exhibit an equation reducing to itself by construction or a fitted parameter renamed as a prediction. Accordingly, the appropriate circularity finding is no significant circularity, with score 0.

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

The central empirical claim rests on the quality and partitioning of the NASA POWER dataset and on default model settings that are not fully specified. The most consequential assumption is that the train/test split is out-of-sample, which is not stated.

free parameters (5)
  • KAN default parameters = not specified (defaults from ref [34])
    The KAN architecture uses default grid size, spline order, and network widths; these choices affect the reported R2 values and are not audited.
  • RNN hidden sizes and layer counts = not reported
    The depth and width of LSTM, GRU, BiLSTM, and BiGRU models are not stated, so the comparison is not fully defined.
  • Sliding window length = 14 days
    The choice of 14 days of observations to predict the next day is arbitrary and could influence the results.
  • Training hyperparameters = lr=0.001, batch_size=64, dropout=0.2 for RNNs
    These settings are chosen by hand and are not justified or searched.
  • Train/validation/test split ratio = 72/8/20
    The split ratio is chosen without explanation, and the temporal ordering of the split is not specified.
assumptions (4)
  • domain assumption NASA POWER reanalysis data are accurate station-level observations for Abidjan and Kigali.
    The paper relies on NASA POWER data (footnote 1) as ground truth without independent validation.
  • domain assumption The MinMaxScaler is fit on the training data only.
    Standard practice, but the paper does not state this; fitting on the full dataset would leak test statistics.
  • domain assumption A 14-day sliding window captures the relevant temporal dynamics for next-day prediction.
    No analysis supports this window length; it is an arbitrary modeling choice.
  • standard math The Kolmogorov-Arnold representation theorem justifies the spline-based network architecture.
    The theorem guarantees representation of continuous functions but does not guarantee learnability or generalization on this data.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Localized Weather Prediction Using Kolmogorov-Arnold Network-Based Models and Deep RNNs." pith.science (2026). https://pith.science/paper/7EG667DM

@misc{pith2026250522686,
  author       = {Pith},
  title        = {Pith review of: Localized Weather Prediction Using Kolmogorov-Arnold Network-Based Models and Deep RNNs},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/7EG667DM}},
  note         = {Machine review of arXiv:2505.22686}
}
abstract

Weather forecasting is crucial for managing risks and economic planning, particularly in tropical Africa, where extreme events severely impact livelihoods. Yet, existing forecasting methods often struggle with the region's complex, non-linear weather patterns. This study benchmarks deep recurrent neural networks such as $\texttt{LSTM, GRU, BiLSTM, BiGRU}$, and Kolmogorov-Arnold-based models $(\texttt{KAN} and \texttt{TKAN})$ for daily forecasting of temperature, precipitation, and pressure in two tropical cities: Abidjan, Cote d'Ivoire (Ivory Coast) and Kigali (Rwanda). We further introduce two customized variants of $ \texttt{TKAN}$ that replace its original $\texttt{SiLU}$ activation function with $ \texttt{GeLU}$ and \texttt{MiSH}, respectively. Using station-level meteorological data spanning from 2010 to 2024, we evaluate all the models on standard regression metrics. $\texttt{KAN}$ achieves temperature prediction ($R^2=0.9986$ in Abidjan, $0.9998$ in Kigali, $\texttt{MSE} < 0.0014~^\circ C ^2$), while $\texttt{TKAN}$ variants minimize absolute errors for precipitation forecasting in low-rainfall regimes. The customized $\texttt{TKAN}$ models demonstrate improvements over the standard $\texttt{TKAN}$ across both datasets. Classical \texttt{RNNs} remain highly competitive for atmospheric pressure ($R^2 \approx 0.83{-}0.86$), outperforming $\texttt{KAN}$-based models in this task. These results highlight the potential of spline-based neural architectures for efficient and data-efficient forecasting.

Figures

Figures reproduced from arXiv: 2505.22686 by the authors.

Figure 2
Figure 2. Abidjan City [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Target Variables Distribution: Row 1 is for Kigali city and row 2 for Abidjan city [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Comparison of predicted vs. actual temperature values in Abidjan using the best [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figures from the paper (9 more)
Figure 5
Figure 5. Figure 5: Comparison of predicted vs. actual temperature values in Kigali using the best [PITH_FULL_IMAGE:figures/full_fig_p009_5.png]
Figure 6
Figure 6. Figure 6: Abidjan and Kigali LSTM [PITH_FULL_IMAGE:figures/full_fig_p015_6.png]
Figure 7
Figure 7. Figure 7: Abidjan and Kigali BiLTM 15 [PITH_FULL_IMAGE:figures/full_fig_p015_7.png]
Figure 8
Figure 8. Figure 8: Abidjan and Kigali GRU [PITH_FULL_IMAGE:figures/full_fig_p016_8.png]
Figure 9
Figure 9. Figure 9: Abidjan and Kigali BiGru [PITH_FULL_IMAGE:figures/full_fig_p016_9.png]
Figure 10
Figure 10. Figure 10: Abidjan and Kigali Ensemble model 16 [PITH_FULL_IMAGE:figures/full_fig_p016_10.png]
Figure 11
Figure 11. Figure 11: Abidjan and Kigali KAN [PITH_FULL_IMAGE:figures/full_fig_p017_11.png]
Figure 12
Figure 12. Figure 12: Abidjan TKAN [PITH_FULL_IMAGE:figures/full_fig_p017_12.png]
Figure 13
Figure 13. Figure 13: Kigali TKAN 17 [PITH_FULL_IMAGE:figures/full_fig_p017_13.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

50 extracted references · 42 canonical work pages

  1. [1]

    Climate forecasting by bidirectional recurrent neural networks

    Sepehr Abbaspour and Hedieh Sajedi. Climate forecasting by bidirectional recurrent neural networks

    work page

  2. [2]

    On the use of kolmogorov–arnold networks for adapting wind numerical weather fore- casts with explainability and interpretability: Application to madeira international airport

    D´ ecio Alves, F´ abio Mendon¸ ca, Sheikh Shanawaz Mostafa, and Fernando Morgado-Dias. On the use of kolmogorov–arnold networks for adapting wind numerical weather fore- casts with explainability and interpretability: Application to madeira international airport. Environmental Research Communications, 6(10):105008, 2024

    work page 2024

  3. [3]

    Use of regression techniques to predict hail size and the probability of large hail

    John Billet, Mark DeLisi, Brian G Smith, and Cory Gates. Use of regression techniques to predict hail size and the probability of large hail. Weather and Forecasting, 12(1):154–164, 1997

    work page 1997

  4. [4]

    Describing video with attention-based bidirectional lstm

    Yi Bin, Yang Yang, Fumin Shen, Ning Xie, Heng Tao Shen, and Xuelong Li. Describing video with attention-based bidirectional lstm. IEEE transactions on cybernetics , 49(7):2631–2641, 2018

    work page 2018

  5. [5]

    The problem of weather prediction, considered from the viewpoints of mechanics and physics

    Vilhelm Bjerknes, Esther Volken, and S Bronnimann. The problem of weather prediction, considered from the viewpoints of mechanics and physics. Meteorologische Zeitschrift, 18(6):663, 2009

    work page 2009

  6. [6]

    Machine learning in weather prediction and climate analyses—applications and perspectives

    Bogdan Bochenek and Zbigniew Ustrnul. Machine learning in weather prediction and climate analyses—applications and perspectives. Atmosphere, 13(2):180, 2022

    work page 2022

  7. [7]

    Probabilistic forecasts of precipitation in terms of quantiles using nwp model output

    John Bjørnar Bremnes. Probabilistic forecasts of precipitation in terms of quantiles using nwp model output. Monthly Weather Review , 132(1):338 – 347, 2004

    work page 2004

  8. [8]

    Chaos and weather prediction, 2002 2002

    Roberto Buizza. Chaos and weather prediction, 2002 2002

    work page 2002

Show all 50 references
  1. [9]

    Weather forecasting for weather derivatives

    Sean D Campbell and Francis X Diebold. Weather forecasting for weather derivatives. Journal of the American Statistical Association , 100(469):6–16, 2005

    2005

  2. [10]

    J. G. Charney, R. Fj ¨Ortoft, and J. Von Neumann. Numerical integration of the barotropic vorticity equation. Tellus, 2(4):237–254, 1950

    1950

  3. [11]

    Data prediction model in wireless sensor networks based on bidirectional lstm

    Hongju Cheng, Zhe Xie, Leihuo Wu, Zhiyong Yu, and Ruixing Li. Data prediction model in wireless sensor networks based on bidirectional lstm. EURASIP Journal on Wireless Communications and Networking , 2019(1):1–12, 2019

    2019

  4. [12]

    Deep blstm- gru model for monthly rainfall prediction: A case study of simtokha, bhutan

    Manoj Chhetri, Sudhanshu Kumar, Partha Pratim Roy, and Byung-Gyu Kim. Deep blstm- gru model for monthly rainfall prediction: A case study of simtokha, bhutan. Remote sensing, 12(19):3174, 2020

    2020

  5. [13]

    On the properties of neural machine translation: Encoder–decoder approaches

    Kyunghyun Cho, Bart van Merri¨ enboer, Dzmitry Bahdanau, and Yoshua Bengio. On the properties of neural machine translation: Encoder–decoder approaches. In Dekai Wu, Marine Carpuat, Xavier Carreras, and Eva Maria Vecchi, editors, Proceedings of SSST-8, Eighth Workshop on Synta...

    2014

  6. [14]

    Socio-economic impacts of extreme weather events in southern africa

    Vincent Davis-Reddy and Mambo. Socio-economic impacts of extreme weather events in southern africa. Climate Risk and Vulnerability: A Handbook for Southern Africa, pp. 30-47 , 2017

    2017

  7. [15]

    Sigmoid-weighted linear units for neural network function approximation in reinforcement learning, 2017

    Stefan Elfwing, Eiji Uchibe, and Kenji Doya. Sigmoid-weighted linear units for neural network function approximation in reinforcement learning, 2017. 12

    2017

  8. [16]

    A revolutionary neural network architecture with interpretability and flexibility based on kolmogorov–arnold for solar radiation and temperature forecasting

    Yuan Gao, Zehuan Hu, Wei-An Chen, Mingzhe Liu, and Yingjun Ruan. A revolutionary neural network architecture with interpretability and flexibility based on kolmogorov–arnold for solar radiation and temperature forecasting. Applied Energy, 378:124844, 2025

    2025

  9. [17]

    Tkan: Temporal kolmogorov-arnold networks

    Remi Genet and Hugo Inzirillo. Tkan: Temporal kolmogorov-arnold networks. arXiv preprint arXiv:2405.07344, 2024

    2024 arXiv

  10. [18]

    Givental, Boris A

    Alexander B. Givental, Boris A. Khesin, Jerrold E. Marsden, Alexander N. Varchenko, Victor A. Vassiliev, Oleg Ya. Viro, and Vladimir M. Zakalyukin, editors.On the representation of functions of several variables as a superposition of functions of a smaller number of variables,...

    2009

  11. [19]

    Raftery, Anton H

    Tilmann Gneiting, Adrian E. Raftery, Anton H. Westveld, and Tom Goldman. Calibrated probabilistic forecasting using ensemble model output statistics and minimum crps estimation. Monthly Weather Review , 133(5):1098 – 1118, 2005

    2005

  12. [20]

    Deep learning for weather forecasting: A cnn-lstm hybrid model for predicting historical temperature data

    Yuhao Gong, Yuchen Zhang, Fei Wang, and Chi-Han Lee. Deep learning for weather forecasting: A cnn-lstm hybrid model for predicting historical temperature data. arXiv preprint arXiv:2410.14963, 2024

    2024 arXiv

  13. [21]

    The climate of niwot ridge, front range, colorado, usa

    David Greenland. The climate of niwot ridge, front range, colorado, usa. Arctic and Alpine Research, 21(4):380–391, 1989

    1989

  14. [22]

    Arma techniques for modelling tree-ring response to climate and for reconstruct- ing variations of paleoclimates

    Joel Guiot. Arma techniques for modelling tree-ring response to climate and for reconstruct- ing variations of paleoclimates. Ecological modelling, 33(2-4):149–171, 1986

    1986

  15. [23]

    Monthly climate prediction using deep convolutional neural network and long short-term memory

    Qingchun Guo, Zhenfang He, and Zhaosheng Wang. Monthly climate prediction using deep convolutional neural network and long short-term memory. Scientific Reports, 14(1):17748, 2024

    2024

  16. [24]

    On typical range, sensitivity, and normalization of mean squared error and nash-sutcliffe efficiency type metrics

    Hoshin Vijai Gupta and Harald Kling. On typical range, sensitivity, and normalization of mean squared error and nash-sutcliffe efficiency type metrics. Water Resources Research, 47(10), 2011

    2011

  17. [25]

    Machine learning-based rainfall prediction: Unveiling insights and forecasting for improved preparedness

    Md Mehedi Hassan, Mohammad Abu Tareq Rony, Md Asif Rakib Khan, Md Mahedi Hassan, Farhana Yasmin, Anindya Nag, Tazria Helal Zarin, Anupam Kumar Bairagi, Samah Alshathri, and Walid El-Shafai. Machine learning-based rainfall prediction: Unveiling insights and forecasting for impr...

    2023

  18. [26]

    Gaussian error linear units (gelus), 2023

    Dan Hendrycks and Kevin Gimpel. Gaussian error linear units (gelus), 2023

    2023

  19. [27]

    Long short-term memory.Neural computation, 9:1735–80, 12 1997

    Sepp Hochreiter and J¨ urgen Schmidhuber. Long short-term memory.Neural computation, 9:1735–80, 12 1997

    1997

  20. [28]

    R. G. Isaacs, R. N. Hoffman, and L. D. Kaplan. Satellite remote sensing of meteorological parameters for global numerical weather prediction. Reviews of Geophysics , 24(4):701–743, 1986

    1986

  21. [29]

    Africa: Types of reported natural disasters, Feb 2022

    Lars Kamer. Africa: Types of reported natural disasters, Feb 2022

    2022

  22. [30]

    W ASHINGTON

    AKIRA KASAHARA and W ARREN M. W ASHINGTON. Ncar global general circulation model of the atmosphere. Monthly Weather Review , 95(7):389 – 402, 1967

    1967

  23. [31]

    A new approach for coupling a limited area model to a gcm for regional climate simulations

    Hideji Kida, Takashi Koide, Hidetaka Sasaki, and Masaru Chiba. A new approach for coupling a limited area model to a gcm for regional climate simulations. Journal of the Meteorological Society of Japan. Ser. II , 69(6):723–728, 1991. 13

    1991

  24. [32]

    Objective prediction of five-day mean temperatures during winter

    William H Klein, Billy M Lewis, and Isadore Enger. Objective prediction of five-day mean temperatures during winter. Journal of Atmospheric Sciences , 16(6):672–682, 1959

    1959

  25. [33]

    Kolmogorov–arnold neural interpolator for downscaling and correcting meteorological fields from in-situ observations

    Zili Liu, Hao Chen, Lei Bai, Wenyuan Li, Wanli Ouyang, Zhengxia Zou, and Zhenwei Shi. Kolmogorov–arnold neural interpolator for downscaling and correcting meteorological fields from in-situ observations. arXiv preprint arXiv:2501.14404 , 2025

    2025 arXiv

  26. [34]

    Kan: Kolmogorov-arnold networks

    Ziming Liu, Yixuan Wang, Sachin Vaidya, Fabian Ruehle, James Halverson, Marin Soljaˇ ci´ c, Thomas Y Hou, and Max Tegmark. Kan: Kolmogorov-arnold networks. arXiv preprint arXiv:2404.19756, 2024

    2024 arXiv

  27. [35]

    Mish: A self regularized non-monotonic activation function, 2020

    Diganta Misra. Mish: A self regularized non-monotonic activation function, 2020

    2020

  28. [36]

    Rainfall variability over multiple cities of india: analysis and forecasting using deep learning models

    Jagabandhu Panda, Nistha Nagar, Asmita Mukherjee, Saugat Bhattacharyya, and Sanjeev Singh. Rainfall variability over multiple cities of india: analysis and forecasting using deep learning models. Earth Science Informatics , 17(2):1105–1124, 2024

    2024

  29. [37]

    A survey on time-series distance measures

    John Paparrizos, Haojun Li, Fan Yang, Kaize Wu, Jens E d’Hondt, and Odysseas Papapetrou. A survey on time-series distance measures. arXiv preprint arXiv:2412.20574 , 2024

    2024 arXiv

  30. [38]

    Pytorch: An imperative style, high-performance deep learning library

    Adam Paszke, Sam Gross, Francisco Massa, Adam Lerer, James Bradbury, Gregory Chanan, Trevor Killeen, Zeming Lin, Estelle Tompson, Noel Gribonval, et al. Pytorch: An imperative style, high-performance deep learning library. In Advances in Neural Information Processing Systems, ...

    2019

  31. [39]

    Phillips

    Norman A. Phillips. The general circulation of the atmosphere: A numerical experiment. Quarterly Journal of the Royal Meteorological Society , 82(352):123–164, 1956

    1956

  32. [40]

    A new paradigm for parameterizations in numerical weather prediction and other atmospheric models

    Roger A Pielke Sr, Toshihisa Matsui, Giovanni Leoncini, Timothy Nobis, Udaysankar S Nair, Er Lu, Joseph Eastman, Sujay Kumar, Christa D Peters-Lidard, Yudong Tian, et al. A new paradigm for parameterizations in numerical weather prediction and other atmospheric models. Nationa...

    2006

  33. [41]

    Deep learning-based weather prediction: a survey

    Xiaoli Ren, Xiaoyong Li, Kaijun Ren, Junqiang Song, Zichen Xu, Kefeng Deng, and Xiang Wang. Deep learning-based weather prediction: a survey. Big Data Research, 23:100178, 2021

    2021

  34. [42]

    A comparative analysis of univariate deep learning-based time-series models for temperature forecasting of the bhubaneshwar

    Naba Krushna Sabat, Rashmiranjan Nayak, Umesh Chandra Pati, and Santos Kumar Das. A comparative analysis of univariate deep learning-based time-series models for temperature forecasting of the bhubaneshwar. In 2022 IEEE 2nd International Symposium on Sustainable Energy, Signal...

    2022

  35. [43]

    Kan-enhanced lstm for accurate and scalable flood forecasting: A case study of the mahanadi basin

    Somrita Sarkar, Anamika Dey, Chandranath Chatterjee, and Pabitra Mitra. Kan-enhanced lstm for accurate and scalable flood forecasting: A case study of the mahanadi basin. In EGU General Assembly 2025 , 2025. doi: 10.5194/egusphere-egu25-8031

    2025 doi

  36. [44]

    Statistical prediction of air pollution levels using non-physical models

    Yoshikazu Sawaragi, Takashi Soeda, H Tamura, Toshio Yoshimura, Sunichiro Ohe, Yoshiteru Chujo, and Hiroichi Ishihara. Statistical prediction of air pollution levels using non-physical models. Automatica, 15(4):441–451, 1979

    1979

  37. [45]

    Yuanhang Su and C.-C. Jay Kuo. On extended long short-term memory and dependent bidirectional recurrent neural network. Neurocomputing, 356:151–161, September 2019

    2019

  38. [46]

    Autoregressive conditional heteroscedasticity in daily wind speed measurements

    RSJ Tol. Autoregressive conditional heteroscedasticity in daily wind speed measurements. Theoretical and applied climatology, 56:113–122, 1997. 14

    1997

  39. [47]

    Artificial intelligence and numerical weather prediction models: A technical survey

    Muhammad Waqas, Usa Wannasingha Humphries, Bunthid Chueasa, and Angkool Wang- wongchai. Artificial intelligence and numerical weather prediction models: A technical survey. Natural Hazards Research, 2024

    2024

  40. [48]

    Cleveland abbe and american meteorology, 1871–

    Edmund P Willis and William H Hooke. Cleveland abbe and american meteorology, 1871–

  41. [49]

    Xueli Zhang, Cankun Zhong, Jianjun Zhang, Ting Wang, and Wing W.Y. Ng. Robust recurrent neural networks for time series forecasting. Neurocomputing, 526:143–157, 2023. Appendix This appendix offers a visual exploration of the forecasting outcomes, presenting a series of images...

    2023

  42. [1901]

    Bulletin of the American Meteorological Society , 87(3):315–326, 2006

    2006

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.