Pith. sign in

REVIEW 5 major objections 4 minor 69 references

Physics-informed deep learning for infectious disease forecasting

T0 review · 5 major / 4 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read A physics-informed neural network that enforces a nine-compartment epidemic model forecasts California COVID-19 outcomes up to four weeks ahead, beating data-only deep learning baselines.

desk verdict A workmanlike incremental PINN forecasting study; the empirical results are plausible but hinge on a potentially non-causal moving average and a partly circular GISST baseline. read the letter →

arxiv 2501.09298 v2 pith:L5N6L7XS submitted 2025-01-16 cs.LG q-bio.QM

Ying Qian , Kui Zhang , Éric Marty , Avranil Basu , Eamon B. O'Dea , Xianqiao Wang , Spencer Fox , Pejman Rohani
show 2 more authors
John M. Drake He Li
This is my paper · ORCID
classification cs.LGq-bio.QM MSC 68T0792D30
keywords physics-informedneuralnetworksinfectiousdiseaseforecastingcompartmentalmodelsCOVID-19time-varyingtransmissionrateprobabilisticCalifornia
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper proposes forecasting weekly COVID-19 cases, deaths, and hospitalizations in California with a physics-informed neural network (PINN): a neural network whose training loss includes a term that enforces the ordinary differential equations of a nine-compartment epidemic model. The authors' claim is that this ODE constraint regularizes the network, so its 1- to 4-week-ahead forecasts beat a naive baseline and several sequence deep learning models (RNN, LSTM, GRU, Transformer) while matching a more elaborate statistical state-space model. A separate sub-network takes time as input and outputs mobility, cumulative vaccine doses, and a time-varying transmission rate, letting covariates enter the mechanistic equations directly. If the claim holds, compact and easy-to-implement PINNs are a viable tool for real-time outbreak forecasting without a separate parameter-inference step.

What carries the argument

The central mechanism is the physics-informed loss: a weighted sum of a data term on observable outcomes and covariates, an initial-condition term, and a residual term enforcing the nine ODEs of the compartmental model. The lower sub-network maps time to mobility, cumulative vaccine doses, and $\beta_t$, so the transmission rate is not a fitted constant but a learned time-varying function that couples the covariates into the ODE system. This is the piece that makes trajectories epidemiologically plausible and prevents the network from fitting noise.

What would settle it

Re-run the same rolling-window protocol on California data that include the Omicron wave (weeks after the paper's cutoff), and check whether the 1-week-ahead MASE for cases stays below 1.0; if it rises above the naive baseline, the fixed-parameter compartmental structure does not generalize across variant regimes. A complementary check is to withhold the mobility and vaccine covariate inputs and retrain: if accuracy is unchanged, the covariate sub-network is not the source of the claimed gains.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that embedding the nine-compartment ODE model of reference [22] into the loss of a two-part neural network produces stable, accurate weekly forecasts. The upper sub-network outputs the nine state variables—susceptible, latent, infectious, diagnosed-not-yet-reported, reported cases, hospitalized, new admissions, deaths-not-yet-reported, and reported deaths—with reported cases, hospitalizations, and reported deaths supervised by data; the lower sub-network outputs mobility, cumulative vaccine doses, and the transmission rate $\beta_t$. The total loss is $L_{\text{data}} + w_{\text{ODE}}L_{\text{ODE}}$, where the ODE loss penalizes residuals of the nine equations and the initial conditions. Trained on California data from July 2020 to April 2022 with a weekly rolling window, the model reports MASE values at or below 1.0 for every target and horizon, and scaled WIS generally below 1.0, with the single exception of 1-week-ahead death quantiles. Compared with a purely data-driven network, the ODE constraint lowers error on nearly all horizons; compared with sequence models, it reduces fluctuation and overfitting; compared with a sophisticated Gaussian state-space forecast, it is competitive and structurally simpler.

Load-bearing premise

The load-bearing assumption is that the nine-compartment structure taken from reference [22], with its fixed parameter values (including incubation and removal rates of 0.25 per day), and with only mobility and cumulative vaccine doses driving the time-varying transmission rate, adequately represents California's infection, reporting, hospitalization, and death processes in the period evaluated.

Editorial extensions

If this is right

  • Weekly 1- to 4-week-ahead forecasts for cases, deaths, and hospitalizations can be produced from one compact architecture with no separate parameter-inference step.
  • The ODE constraint acts as a regularizer: dropping it (the pure NN ablation) raises MASE and WIS at nearly every horizon, and removing L2 regularization degrades cases by 22.48%, deaths by 9.16%, and hospitalizations by 42.54%.
  • The method stays competitive with a Gaussian state-space model while using a simpler structure, so it is a practical option for jurisdictions that lack elaborate inference pipelines.
  • Because training length beyond about 16 weeks gives little benefit, forecasts can be updated on a rolling window of recent data, which suits non-stationary outbreaks.
  • Quantile forecasts with a fixed Gaussian error assumption give intervals that widen with horizon, matching the typical demand for probabilistic epidemic forecasts.

Reading between the lines

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

  • A natural extension the paper leaves untested is transferring the same architecture to other states or pathogens; the fixed parameter set and covariate definitions would need re-estimation, and that re-estimation could be checked against held-out regions.
  • The comparison with the state-space model is not uniform: the PINN wins on hospitalizations while the state-space model wins on cases and point forecasts of deaths. That asymmetry suggests per-target weighting of the ODE loss could push the method further, but the paper does not explore it.
  • The single-constant-Gaussian approximation of predictive uncertainty is a simplification; replacing it with ensemble or Bayesian posterior over network weights could improve quantile coverage at longer horizons, a direction the paper itself flags.
  • Because the paper deliberately stops before the Omicron-dominated period, the strongest test of the claim would be an out-of-sample run on later variant waves; the paper's own training-length analysis hints that a short rolling window would adapt, but that is my inference, not its result.
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

5 major / 4 minor

Summary. The manuscript proposes a physics-informed neural network (PINN) for weekly COVID-19 forecasting in California. The model embeds the nine-compartment ODE model of O'Dea and Drake into the loss function, while a second sub-network maps time to mobility, cumulative vaccine doses, and transmission rate. The method is evaluated in a rolling-window design for 1-4 week-ahead forecasts of cases, deaths, and hospitalizations, and is compared with a data-only neural network, the GISST model, and RNN/LSTM/GRU/Transformer baselines using MASE and WIS. The central claim is that this compact PINN outperforms naive and sequence-learning baselines and is comparable to or better than GISST, while being simpler to implement.

Significance. If the claims hold, the paper provides a useful demonstration that a compact PINN with a nine-compartment ODE constraint can produce competitive weekly COVID-19 forecasts. Strengths include the use of standardized Forecast Hub targets and metrics, the inclusion of several baselines (naive, data-only NN, GISST, and four sequence models), and ablation studies for training-window length and L2 regularization. The authors also state that code and data are publicly available. At present, however, the quantitative comparison in Table 3 is not fully validated because of unresolved preprocessing, uncertainty-reporting, and statistical-support issues; the contribution is a promising proof-of-concept rather than a definitive benchmark.

major comments (5)
  1. [§2.1.2, §2.2, Eq. (2)-(5)] Section 2.1.2 states that the daily series are preprocessed by 'computing a 7-day moving average and aggregating the daily values into a weekly time series,' but it does not state that the moving average is trailing. Under the rolling-window protocol of Section 2.2, training data are available only up to the forecast origin. If the implementation uses a centered moving average (e.g., pandas rolling(7, center=True) or np.convolve with 'same'), the smoothed value at the last training time uses observations from the following three days; those observations enter the data loss in Eq. (2), the initial-condition loss in Eq. (4), and the ODE residual in Eq. (5). This would make the 1-week-ahead forecasts partly in-sample and would affect all horizons through the ODE state, invalidating the comparison in Table 3. The authors should state explicitly that the moving average is causal/trailing and confirm this in the released code; if it is centered, the experiments must be rerun.
  2. [§2.1.2] The evaluation period is restricted by design: the first 20 weeks of hospitalization data are excluded, and all data after week 110 are discarded to 'maintain focus on well-characterized dynamics' (the two major Beta/Delta peaks). This is a post-hoc selection of the evaluation window, and it means the favorable results in Table 3 are established only for two variant waves, not for the full California record. The general claim in the abstract and Section 4 that the model 'accurately predicts cases, deaths, and hospitalizations' should be tempered to this selected period, or the authors should provide results on the full available record as a robustness check.
  3. [§2.3, Table 3] MASE and WIS are reported as single point estimates for each method, horizon, and target, with no standard errors, confidence intervals, or significance tests. The neural-network methods are stochastic, and several differences are modest (for example, 1-week-ahead death MASE is 1.00 for PINN versus 0.88 for GISST, and 2-week cases MASE is 0.77 for PINN versus 0.75 for GISST). As written, the claims in Section 3.3 that PINNs 'consistently outperform' the alternatives, or are 'comparable' to GISST, are not statistically supported. Please provide multiple-seed results with interval estimates, or at least a bootstrapped assessment of the score differences.
  4. [§2.3] The WIS computation assumes a Gaussian predictive distribution with a constant standard deviation σ, described as 'the standard deviation of all prediction errors.' The manuscript does not state how σ is obtained or whether it is fixed before seeing the evaluation period. If σ is computed from the full set of forecast errors over the evaluation period, the quantile forecasts are not out-of-sample and the WIS values in Table 3 are not a valid comparison. Please specify the estimation procedure (for example, from training residuals at each forecast origin) and confirm that it is causal.
  5. [§2.2, Table 2] The ODE constraint is imported with fixed parameter values from O'Dea and Drake [22] (η=0.25, γ=0.25, etc.), while only β_t, p_h, and p_d are estimated. These fixed values are not re-validated for the California Beta/Delta waves. Because the ODE loss acts as a regularizer in every training run, a misspecified fixed parameter could bias the forecasts. A sensitivity analysis over the fixed parameters (or a version of the model that estimates them) would address this concern and is well within the scope of the paper.
minor comments (4)
  1. [§2.4, Fig. 3 caption] The GISST baseline is cited as [23] in Section 2.4 but as [22] in the caption of Fig. 3; please reconcile the references and state clearly which published model is used.
  2. [Data Availability] The repository URL is given as 'https://github.com/dpdclub/PINNs-for-Epidemiology- ' with a trailing hyphen; this appears truncated and should be corrected.
  3. [§2.3] The quantity in Eq. (7) is defined as the ratio of the model MAE to the naive model MAE. This is a relative MAE, not the standard MASE defined in [61]; please use a less misleading name or explicitly define the deviation from the standard definition.
  4. [§2.2] The ODE weight w_ODE and the L2 coefficient were 'manually tuned to optimize predictive accuracy'; a brief sensitivity analysis over w_ODE would reassure readers that the reported gains do not depend on a single hand-picked value.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: held-out forecast evaluation with an externally published compartmental model; no fitted quantity is renamed as a prediction.

full rationale

The paper's central empirical claim is that its PINN model produces accurate 1- to 4-week-ahead forecasts of COVID-19 cases, deaths, and hospitalizations in California. This is evaluated with MASE and WIS against held-out observations in a rolling-window design (Section 2.2 and Table 3), so the forecasts are not fitted values renamed as predictions. The compartmental model and fixed parameter values are adopted from O'Dea and Drake [22]; although the author lists overlap with the present paper, that prior model is an independently published, externally falsifiable modeling choice, not a theorem whose conclusion is the PINN forecast. The GISST baseline solves the same compartmental model, which makes the comparison an implementation and calibration contrast rather than a circular reduction. The Gaussian quantile construction uses a constant sigma derived from prediction errors; this is a standard way to convert point forecasts into intervals and is applied to all compared methods, so it does not define the target as the input. The only notable validity concern is that the 7-day moving average in Section 2.1.2 is not explicitly stated to be trailing; a centered average would leak near-future observations into training. That is a data-leakage and correctness issue, not a circularity of definition or self-citation, and it cannot be established as circular from the text alone.

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

No new physical entities are introduced. The central claim rests on adopting the O'Dea-Drake compartmental model, a set of fitted parameters and hyperparameters, and a simplistic Gaussian uncertainty model. The compartmental model and fixed constants are not re-derived in this paper.

free parameters (6)
  • transmission rate beta_t = time-varying function estimated by lower sub-network
    Key parameter in the ODE system, learned from data; controls infection flow.
  • probability of hospitalization p_h = estimated during training
    One of the three parameters estimated by minimizing the PINN loss; affects H and A compartments.
  • probability of death given hospitalization p_d = estimated during training
    Estimated parameter affecting D and Dr compartments.
  • quantile spread sigma = standard deviation of all prediction errors, constant in time
    Used to construct Gaussian quantile forecasts; fitted from model errors, not derived from an independent source.
  • ODE loss weight w_ODE = 0.1
    Manually tuned to balance data and ODE losses; directly affects training and forecasts.
  • L2 regularization coefficient = 1e-5
    Manually tuned; Appendix 5.2 shows removing it degrades accuracy.
assumptions (4)
  • domain assumption The O'Dea-Drake nine-compartment model with eleven parameters correctly describes COVID-19 dynamics in California.
    The ODE system is the 'physics' constraint in the PINN; the paper adopts it wholesale from [22] without independent validation.
  • domain assumption Fixed parameter values from [22] (incubation rate 0.25, removal rate 0.25, reporting rates, etc.) are appropriate for the studied period.
    Table 2 lists these constants; no sensitivity analysis or estimation is performed for them.
  • ad hoc to paper The predictive distribution is Gaussian with a constant standard deviation.
    Section 2.3 assumes all prediction errors are Gaussian with fixed spread; this is not validated against observed forecast errors.
  • domain assumption Unobserved latent initial conditions for the nine state variables are set consistently with observed data.
    The model trains with an initial condition loss, but the actual initial values for latent compartments are not reported.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Physics-informed deep learning for infectious disease forecasting." pith.science (2026). https://pith.science/paper/L5N6L7XS

@misc{pith2026250109298,
  author       = {Pith},
  title        = {Pith review of: Physics-informed deep learning for infectious disease forecasting},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/L5N6L7XS}},
  note         = {Machine review of arXiv:2501.09298}
}
read the original abstract

Accurate forecasting of contagious diseases is critical for public health policymaking and pandemic preparedness. We propose a new infectious disease forecasting model based on physics-informed neural networks (PINNs), an emerging scientific machine learning approach. By embedding a compartmental model into the loss function, our method integrates epidemiological theory with data, helping to prevent model overfitting. We further enhance the model with a sub-network that accounts for covariates such as mobility and cumulative vaccine doses, which influence the transmission rate. Using state-level COVID-19 data from California, we demonstrate that the PINN model accurately predicts cases, deaths, and hospitalizations, aligning well with existing benchmarks. Notably, the PINN model outperforms naive baseline forecasts and several sequence deep learning models, including Recurrent Neural Networks (RNNs), Long Short-Term Memory (LSTM) networks, Gated Recurrent Units (GRUs), and Transformers. It also achieves performance comparable to a sophisticated Gaussian infection state forecasting model that combines compartmental dynamics, a data observation model, and parameter regression. However, the PINN model features a simpler structure and is easier to implement. In summary, we systematically evaluate the PINN model's ability to forecast infectious disease dynamics, demonstrating its potential as an efficient computational tool to strengthen forecasting capabilities.

Figures

Figures reproduced from arXiv: 2501.09298 by the authors.

Figure 1
Figure 1. A. Schematic of the proposed PINNs model for infectious disease forecasting. The model comprises two sub-networks: the upper sub-network predicts the state variables in the compart￾mental model, while the lower sub-network estimates the time-dependent model parameters. The output u1 represents the nine compartmental state variables, including X, L, Z, Zr, H, A, D, Dr. The output u2 represents factors including mobil… view at source ↗
Figure 2
Figure 2. Original dataset (upper panel) and preprocessed (lower panel) dataset used for training [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. PINNs’ point predictions on the number of cases, deaths, and hospitalizations for the [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: PINNs’ quantile predictions on the number of cases, deaths, and hospitalizations for the [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: A MASE comparison of Recurrent Neural Networks (RNNs), Long Short-Term Memory [PITH_FULL_IMAGE:figures/full_fig_p013_5.png]
Figure 6
Figure 6. Figure 6: MASE comparison with Recurrent Neural Networks (RNNs), Long Short-Term Memory [PITH_FULL_IMAGE:figures/full_fig_p014_6.png]
Figure 7
Figure 7. Figure 7: MASE for the predictions of PINNs models on cases, deaths, and hospitalizations when [PITH_FULL_IMAGE:figures/full_fig_p016_7.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

69 extracted references · 61 canonical work pages

  1. [22]

    A semi-parametric, state-space compartmental model with time-dependent parameters for forecasting covid-19 cases, hospitalizations and deaths

    Eamon B O’Dea and John M Drake. A semi-parametric, state-space compartmental model with time-dependent parameters for forecasting covid-19 cases, hospitalizations and deaths. Journal of the Royal Society Interface , 19(187):20210702, 2022

    work page 2022

  2. [1]

    Provisional mortality data—united states, 2021

    Farida B Ahmad, Jodi A Cisewski, and Robert N Anderson. Provisional mortality data—united states, 2021. Morbidity and Mortality Weekly Report , 71(17):597, 2022

    work page 2021

  3. [2]

    The challenge of emerging and re-emerging infectious diseases

    David M Morens, Gregory K Folkers, and Anthony S Fauci. The challenge of emerging and re-emerging infectious diseases. Nature, 430(6996):242–249, 2004

    work page 2004

  4. [3]

    Climate change increases cross-species viral transmission risk

    Colin J Carlson, Gregory F Albery, Cory Merow, Christopher H Trisos, Casey M Zipfel, Evan A Eskew, Kevin J Olival, Noam Ross, and Shweta Bansal. Climate change increases cross-species viral transmission risk. Nature, 607(7919):555–562, 2022

    work page 2022

  5. [4]

    Improving pandemic response: employing mathematical modeling to confront coronavirus disease 2019

    Matthew Biggerstaff, Rachel B Slayton, Michael A Johansson, and Jay C Butler. Improving pandemic response: employing mathematical modeling to confront coronavirus disease 2019. Clinical Infectious Diseases, 74(5):913–917, 2022

    work page 2019

  6. [5]

    Applying infectious disease forecasting to public health: a path forward using influenza fore- casting examples

    Chelsea S Lutz, Mimi P Huynh, Monica Schroeder, Sophia Anyatonwu, F Scott Dahlgren, Gregory Danyluk, Danielle Fernandez, Sharon K Greene, Nodar Kipshidze, Leann Liu, et al. Applying infectious disease forecasting to public health: a path forward using influenza fore- casting examples. BMC Public Health , 19(1):1–12, 2019

    work page 2019

  7. [6]

    An accurate hierarchical model to forecast diverse seasonal infectious diseases

    B KM Case, Mariah Victoria Salcedo, and Spencer J Fox. An accurate hierarchical model to forecast diverse seasonal infectious diseases. medRxiv, pages 2025–03, 2025

    work page 2025

  8. [7]

    Infectious Diseases of Humans

    R M Anderson and R M May. Infectious Diseases of Humans. Oxford University Press. Oxford University Press, 1991

    work page 1991

Show all 69 references
  1. [8]

    Modelling Infectious Diseases: In Humans and Animals

    M J Keeling and P Rohani. Modelling Infectious Diseases: In Humans and Animals . Princeton University Press. Princeton University Press, 2008

    2008

  2. [9]

    Compartmental models in epidemiology

    Fred Brauer. Compartmental models in epidemiology. Mathematical epidemiology, pages 19– 79, 2008

    2008

  3. [10]

    Mathematical models in epidemiology, volume 32

    Fred Brauer, Carlos Castillo-Chavez, Zhilan Feng, et al. Mathematical models in epidemiology, volume 32. Springer, 2019. 17

    2019

  4. [11]

    An introduction to mathematical epidemiology , volume 61

    Maia Martcheva. An introduction to mathematical epidemiology , volume 61. Springer, 2015

    2015

  5. [12]

    Mathematical modelling and prediction in infectious disease epidemiology

    Amit Huppert and Guy Katriel. Mathematical modelling and prediction in infectious disease epidemiology. Clinical microbiology and infection , 19(11):999–1005, 2013

    2013

  6. [13]

    Appropriate Models for the Manage- ment of Infectious Diseases

    Helen J Wearing, Pejman Rohani, and Matt J Keeling. Appropriate Models for the Manage- ment of Infectious Diseases. PLoS Medicine, 2(7):e174, 2005

    2005

  7. [14]

    Structural identifiability and observability of compartmental models of the covid-19 pandemic

    Gemma Massonis, Julio R Banga, and Alejandro F Villaverde. Structural identifiability and observability of compartmental models of the covid-19 pandemic. Annual reviews in control , 51:441–459, 2021

    2021

  8. [15]

    A fractional-order compartmental model for the spread of the covid-19 pandemic

    Toheeb A Biala and AQM Khaliq. A fractional-order compartmental model for the spread of the covid-19 pandemic. Communications in Nonlinear Science and Numerical Simulation , 98:105764, 2021

    2021

  9. [16]

    Mathematical modeling of covid-19 transmission dynamics with a case study of wuhan

    Fa ¨ ı¸ cal Nda ¨ ırou, Iv´ an Area, Juan J Nieto, and Delfim FM Torres. Mathematical modeling of covid-19 transmission dynamics with a case study of wuhan. Chaos, Solitons & Fractals , 135:109846, 2020

    2020

  10. [17]

    Compartmental models of the covid-19 pandemic for physicians and physician-scientists

    Anas Abou-Ismail. Compartmental models of the covid-19 pandemic for physicians and physician-scientists. SN comprehensive clinical medicine , 2(7):852–858, 2020

    2020

  11. [18]

    Sirsi compartmental model for covid-19 pandemic with immunity loss

    Cristiane M Batistela, Diego PF Correa, ´Atila M Bueno, and Jos´ e Roberto C Piqueira. Sirsi compartmental model for covid-19 pandemic with immunity loss. Chaos, Solitons & Fractals , 142:110388, 2021

    2021

  12. [19]

    New compartment model for covid-19

    Takashi Odagaki. New compartment model for covid-19. Scientific Reports, 13(1):5409, 2023

    2023

  13. [20]

    Matthieu Domenech de Cell` es, Felicia M. G. Magpantay, Aaron A. King, and Pejman Ro- hani. The impact of past vaccination coverage and immunity on pertussis resurgence. Science Translational Medicine, 10(434):eaaj1748, 03 2018

    2018

  14. [21]

    Measuring the impact of Ebola control measures in Sierra Leone

    Adam J Kucharski, Anton Camacho, Stefan Flasche, Rebecca E Glover, W John Edmunds, and Sebastian Funk. Measuring the impact of Ebola control measures in Sierra Leone. Proceedings of the National Academy of Sciences of the United States of America , 112(46):14366 – 14371, 11 2015

    2015

  15. [23]

    A data-driven semi-parametric model of sars-cov-2 trans- mission in the united states

    John M Drake, Andreas Handel, ´Eric Marty, Eamon B O’Dea, Tierney O’Sullivan, Giovanni Righi, and Andrew T Tredennick. A data-driven semi-parametric model of sars-cov-2 trans- mission in the united states. PLOS Computational Biology , 19(11):e1011610, 2023

    2023

  16. [24]

    Real-time mechanistic bayesian forecasts of covid-19 mortality

    Graham C Gibson, Nicholas G Reich, and Daniel Sheldon. Real-time mechanistic bayesian forecasts of covid-19 mortality. medRxiv, 2020

    2020

  17. [25]

    Deep learning for epidemio- logical predictions

    Yuexin Wu, Yiming Yang, Hiroshi Nishiura, and Masaya Saitoh. Deep learning for epidemio- logical predictions. In The 41st International ACM SIGIR Conference on Research & Devel- opment in Information Retrieval , pages 1085–1088, 2018. 18

    2018

  18. [26]

    Deep learning for few-shot white blood cell image classification and feature learning

    Yixiang Deng and He Li. Deep learning for few-shot white blood cell image classification and feature learning. Computer Methods in Biomechanics and Biomedical Engineering: Imaging & Visualization , 11(6):2081–2091, 2023

    2023

  19. [27]

    A systematic study of the performance of machine learning models on analyzing the association between semen quality and environmental pollutants

    Lu Lu, Ying Qian, Yihang Dong, Han Su, Yunxin Deng, Qiang Zeng, and He Li. A systematic study of the performance of machine learning models on analyzing the association between semen quality and environmental pollutants. Frontiers in Physics, 11:1259273, 2023

    2023

  20. [28]

    A deep learning model for efficient end-to-end stratification of thrombotic risk in left atrial appendage

    Qi Gao, Hongtao Lin, Jianghong Qian, Xingli Liu, Shengze Cai, He Li, Hongguang Fan, and Zhe Zheng. A deep learning model for efficient end-to-end stratification of thrombotic risk in left atrial appendage. Engineering Applications of Artificial Intelligence , 126:107187, 2023

    2023

  21. [29]

    Mathematical models for predicting covid-19 pandemic: a review

    Vishnu Vytla, Sravanth Kumar Ramakuri, Anudeep Peddi, K Kalyan Srinivas, and N Nithish Ragav. Mathematical models for predicting covid-19 pandemic: a review. In Journal of Physics: Conference Series , volume 1797, page 012009. IOP Publishing, 2021

    2021

  22. [30]

    Kevrekidis, Lu Lu, Paris Perdikaris, Sifan Wang, and Liu Yang

    George Em Karniadakis, Ioannis G. Kevrekidis, Lu Lu, Paris Perdikaris, Sifan Wang, and Liu Yang. Physics-informed machine learning. Nature Reviews Physics , 3(6):422–440, 2021

    2021

  23. [31]

    Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations

    Maziar Raissi, Alireza Yazdani, and George Em Karniadakis. Hidden fluid mechanics: Learning velocity and pressure fields from flow visualizations. Science, 367(6481):1026–1030, 2020

    2020

  24. [32]

    Extraction of mechanical properties of materials through deep learning from instru- mented indentation

    Lu Lu, Ming Dao, Punit Kumar, Upadrasta Ramamurty, George Em Karniadakis, and Subra Suresh. Extraction of mechanical properties of materials through deep learning from instru- mented indentation. Proceedings of the National Academy of Sciences , 117(13):7052–7062, 2020

    2020

  25. [33]

    Multi-head physics-informed neural networks for learning functional priors and uncertainty quantification

    Zongren Zou and George Em Karniadakis. Multi-head physics-informed neural networks for learning functional priors and uncertainty quantification. Journal of Computational Physics , page 113947, 2025

    2025

  26. [34]

    Correcting model misspecification in physics-informed neural networks (PINNs)

    Zongren Zou, Xuhui Meng, and George Em Karniadakis. Correcting model misspecification in physics-informed neural networks (PINNs). Journal of Computational Physics , 505:112918, 2024

    2024

  27. [35]

    Discovering a reaction– diffusion model for Alzheimer’s disease by combining PINNs with symbolic regression

    Zhen Zhang, Zongren Zou, Ellen Kuhl, and George Em Karniadakis. Discovering a reaction– diffusion model for Alzheimer’s disease by combining PINNs with symbolic regression. Com- puter Methods in Applied Mechanics and Engineering , 419:116647, 2024

    2024

  28. [36]

    Uncertainty quantification for noisy inputs–outputs in physics-informed neural networks and neural operators

    Zongren Zou, Xuhui Meng, and George Em Karniadakis. Uncertainty quantification for noisy inputs–outputs in physics-informed neural networks and neural operators. Computer Methods in Applied Mechanics and Engineering , 433:117479, 2025

    2025

  29. [37]

    NeuralUQ: A com- prehensive library for uncertainty quantification in neural differential equations and operators

    Zongren Zou, Xuhui Meng, Apostolos F Psaros, and George E Karniadakis. NeuralUQ: A com- prehensive library for uncertainty quantification in neural differential equations and operators. SIAM Review, 66(1):161–190, 2024

    2024

  30. [38]

    B-pinns: Bayesian physics-informed neu- ral networks for forward and inverse PDE problems with noisy data

    Liu Yang, Xuhui Meng, and George Em Karniadakis. B-pinns: Bayesian physics-informed neu- ral networks for forward and inverse PDE problems with noisy data. Journal of Computational Physics, 425:109913, 2021. 19

    2021

  31. [39]

    Ppinn: Parareal physics- informed neural network for time-dependent pdes

    Xuhui Meng, Zhen Li, Dongkun Zhang, and George Em Karniadakis. Ppinn: Parareal physics- informed neural network for time-dependent pdes. Computer Methods in Applied Mechanics and Engineering, 370:113250, 2020

    2020

  32. [40]

    Gradient-enhanced physics- informed neural networks for forward and inverse pde problems

    Jeremy Yu, Lu Lu, Xuhui Meng, and George Em Karniadakis. Gradient-enhanced physics- informed neural networks for forward and inverse pde problems. Computer Methods in Applied Mechanics and Engineering , 393:114823, 2022

    2022

  33. [41]

    Physics-informed neural networks for solving forward and inverse flow problems via the boltzmann-bgk formulation

    Qin Lou, Xuhui Meng, and George Em Karniadakis. Physics-informed neural networks for solving forward and inverse flow problems via the boltzmann-bgk formulation. Journal of Computational Physics , 447:110676, 2021

    2021

  34. [42]

    Deepxde: A deep learning library for solving differential equations

    Lu Lu, Xuhui Meng, Zhiping Mao, and George Em Karniadakis. Deepxde: A deep learning library for solving differential equations. SIAM review, 63(1):208–228, 2021

    2021

  35. [43]

    From PINNs to PIKANs: Recent advances in physics-informed machine learning

    Juan Diego Toscano, Vivek Oommen, Alan John Varghese, Zongren Zou, Nazanin Ah- madi Daryakenari, Chenxi Wu, and George Em Karniadakis. From PINNs to PIKANs: Recent advances in physics-informed machine learning. Machine Learning for Computational Science and Engineering, 1(1):1...

    2025

  36. [44]

    Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations

    Maziar Raissi, Paris Perdikaris, and George E Karniadakis. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational Physics , 378:686–707, 2019

    2019

  37. [45]

    Aoslo-net: a deep learning-based method for automatic segmentation of retinal microaneurysms from adaptive optics scanning laser ophthalmoscopy images

    Qian Zhang, Konstantina Sampani, Mengjia Xu, Shengze Cai, Yixiang Deng, He Li, Jennifer K Sun, and George Em Karniadakis. Aoslo-net: a deep learning-based method for automatic segmentation of retinal microaneurysms from adaptive optics scanning laser ophthalmoscopy images. Tra...

    2022

  38. [46]

    Tgm-nets: A deep learning framework for enhanced forecasting of tumor growth by integrating imaging and modeling

    Qijing Chen, Qi Ye, Weiqi Zhang, He Li, and Xiaoning Zheng. Tgm-nets: A deep learning framework for enhanced forecasting of tumor growth by integrating imaging and modeling. Engineering Applications of Artificial Intelligence , 126:106867, 2023

    2023

  39. [47]

    Transfer learning on physics-informed neural networks for tracking the hemodynamics in the evolving false lumen of dissected aorta

    Mitchell Daneker, Shengze Cai, Ying Qian, Eric Myzelev, Arsh Kumbhat, He Li, and Lu Lu. Transfer learning on physics-informed neural networks for tracking the hemodynamics in the evolving false lumen of dissected aorta. Nexus, 1(2), 2024

    2024

  40. [48]

    A deep neural network for operator learning enhanced by attention and gating mechanisms for long-time forecasting of tumor growth

    Qijing Chen, He Li, and Xiaoning Zheng. A deep neural network for operator learning enhanced by attention and gating mechanisms for long-time forecasting of tumor growth. Engineering with Computers , pages 1–111, 2024

    2024

  41. [49]

    Coagulo-net: Enhancing the mathematical modeling of blood coagulation using physics-informed neural networks

    Ying Qian, Ge Zhu, Zhen Zhang, Susree Modepalli, Yihao Zheng, Xiaoning Zheng, Galit Frydman, and He Li. Coagulo-net: Enhancing the mathematical modeling of blood coagulation using physics-informed neural networks. Neural Networks, page 106732, 2024

    2024

  42. [50]

    Artificial intelligence velocimetry and microaneurysm-on-a-chip for three-dimensional analysis of blood flow in physiology and disease

    Shengze Cai, He Li, Fuyin Zheng, Fang Kong, Ming Dao, George Em Karniadakis, and Subra Suresh. Artificial intelligence velocimetry and microaneurysm-on-a-chip for three-dimensional analysis of blood flow in physiology and disease. Proceedings of the National Academy of Science...

    2021

  43. [51]

    Bayesian physics informed neural networks for real-world nonlinear dynamical systems

    Kevin Linka, Amelie Sch¨ afer, Xuhui Meng, Zongren Zou, George Em Karniadakis, and Ellen Kuhl. Bayesian physics informed neural networks for real-world nonlinear dynamical systems. Computer Methods in Applied Mechanics and Engineering , 402:115346, 2022. 20

    2022

  44. [52]

    Thermodynamically consistent physics-informed neural networks for hyperbolic systems

    Ravi G Patel, Indu Manickam, Nathaniel A Trask, Mitchell A Wood, Myoungkyu Lee, Ignacio Tomas, and Eric C Cyr. Thermodynamically consistent physics-informed neural networks for hyperbolic systems. Journal of Computational Physics , 449:110754, 2022

    2022

  45. [53]

    A physics-informed op- erator regression framework for extracting data-driven continuum models

    Ravi G Patel, Nathaniel A Trask, Mitchell A Wood, and Eric C Cyr. A physics-informed op- erator regression framework for extracting data-driven continuum models. Computer Methods in Applied Mechanics and Engineering , 373:113500, 2021

    2021

  46. [54]

    Learning and discovering mul- tiple solutions using physics-informed neural networks with random initialization and deep ensemble

    Zongren Zou, Zhicheng Wang, and George Em Karniadakis. Learning and discovering mul- tiple solutions using physics-informed neural networks with random initialization and deep ensemble. arXiv preprint arXiv:2503.06320 , 2025

    2025 arXiv

  47. [55]

    Evaluating epidemic forecasts in an interval format

    Johannes Bracher, Evan L Ray, Tilmann Gneiting, and Nicholas G Reich. Evaluating epidemic forecasts in an interval format. PLoS computational biology, 17(2):e1008618, 2021

    2021

  48. [56]

    Ensemble forecasts of coronavirus disease 2019 (covid-19) in the us

    Evan L Ray, Nutcha Wattanachit, Jarad Niemi, Abdul Hannan Kanji, Katie House, Estee Y Cramer, Johannes Bracher, Andrew Zheng, Teresa K Yamana, Xinyue Xiong, et al. Ensemble forecasts of coronavirus disease 2019 (covid-19) in the us. MedRXiv, pages 2020–08, 2020

    2019

  49. [57]

    An interactive web-based dashboard to track covid-19 in real time

    Ensheng Dong, Hongru Du, and Lauren Gardner. An interactive web-based dashboard to track covid-19 in real time. The Lancet infectious diseases , 20(5):533–534, 2020

    2020

  50. [58]

    An open reposi- tory of real-time covid-19 indicators

    Alex Reinhart, Logan Brooks, Maria Jahja, Aaron Rumack, Jingjing Tang, Sumit Agrawal, Wael Al Saeed, Taylor Arnold, Amartya Basu, Jacob Bien, et al. An open reposi- tory of real-time covid-19 indicators. Proceedings of the National Academy of Sciences , 118(51):e2111452118, 2021

    2021

  51. [59]

    Google covid-19 community mobility reports

    Google LLC. Google covid-19 community mobility reports. See https://www.google.com/ covid19/mobility/, 2020

    2020

  52. [60]

    Adam: A method for stochastic optimization

    Diederik P Kingma and Jimmy Ba. Adam: A method for stochastic optimization. arXiv preprint arXiv:1412.6980, 2014

    2014 arXiv

  53. [61]

    Another look at measures of forecast accuracy

    Rob J Hyndman and Anne B Koehler. Another look at measures of forecast accuracy. Inter- national journal of forecasting , 22(4):679–688, 2006

    2006

  54. [62]

    Recurrent neural networks

    Stephen Grossberg. Recurrent neural networks. Scholarpedia, 8(2):1888, 2013

    2013

  55. [63]

    Long short-term memory

    Sepp Hochreiter and J¨ urgen Schmidhuber. Long short-term memory. Neural computation , 9(8):1735–1780, 1997

    1997

  56. [64]

    Gate-variants of gated recurrent unit (gru) neural networks

    Rahul Dey and Fathi M Salem. Gate-variants of gated recurrent unit (gru) neural networks. In 2017 IEEE 60th international midwest symposium on circuits and systems (MWSCAS) , pages 1597–1600. IEEE, 2017

    2017

  57. [65]

    Attention mechanism, transformers, bert, and gpt: tuto- rial and survey

    Benyamin Ghojogh and Ali Ghodsi. Attention mechanism, transformers, bert, and gpt: tuto- rial and survey. 2020

    2020

  58. [66]

    A physics-informed neural network to model covid- 19 infection and hospitalization scenarios

    Sarah Berkhahn and Matthias Ehrhardt. A physics-informed neural network to model covid- 19 infection and hospitalization scenarios. Advances in continuous and discrete models , 2022(1):61, 2022. 21

    2022

  59. [67]

    Einns: epidemiologically-informed neural networks

    Alexander Rodr ´ ıguez, Jiaming Cui, Naren Ramakrishnan, Bijaya Adhikari, and B Aditya Prakash. Einns: epidemiologically-informed neural networks. In Proceedings of the AAAI conference on artificial intelligence , volume 37, pages 14453–14460, 2023

    2023

  60. [68]

    Physics-informed machine learning for the covid-19 pandemic: Adherence to social distancing and short-term predictions for eight coun- tries

    Georgios D Barmparis and Giorgos P Tsironis. Physics-informed machine learning for the covid-19 pandemic: Adherence to social distancing and short-term predictions for eight coun- tries. Quantitative Biology, 10(2):139–149, 2022

    2022

  61. [69]

    Neural networks for endemic measles dynamics: compar- ative analysis and integration with mechanistic models

    Wyatt Madden, Wei Jin, Ben Lopman, Andreas Zufle, Benjamin D Dalziel, Jessica Metcalf, Bryan D Grenfell, and Max SY Lau. Neural networks for endemic measles dynamics: compar- ative analysis and integration with mechanistic models. medRxiv, pages 2024–05, 2024. 22

    2024

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.