Pith. sign in

REVIEW 4 major objections 6 minor 61 references

Towards the efficacy of federated prediction for epidemics on networks

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

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

Pith's one-line read The paper claims that federated learning can deliver effective node-level epidemic forecasts on partitioned networks, with a spatio-temporal graph attention model (STGAT) preferred for fluctuating epidemic dynamics, LSTM sufficient for…

desk verdict Useful empirical sweep of federated epidemic prediction, but the STGAT-versus-LSTM comparison is confounded by model capacity and the evidence base lacks error bars; worth serious review with requested revisions. read the letter →

arxiv 2412.02161 v1 pith:CXVNP6TG submitted 2024-12-03 cs.SI cs.DCcs.LG

Chengpeng Fu , Tong Li , Hao Chen , Wen Du , Zhidong He This is my paper · ORCID
classification cs.SIcs.DCcs.LG
keywords federatedlearningepidemicpredictionspatio-temporalgraphattentionnetworksLSTMefficacyenergypartitioningnetworkepidemiologyFedProx
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

Federated epidemic prediction asks whether multiple data-holding regions can train a shared forecaster without pooling their health data. The paper argues the answer is yes for a broad family of network epidemic models — two-state and multi-state, Markovian and non-Markovian, static and time-varying infection rates — when each region trains locally and a central server aggregates model parameters. It shows that no single architecture wins everywhere: a spatio-temporal graph attention network (STGAT) is better at capturing fluctuating dynamics, while a plain temporal LSTM holds its own on simpler outbreak patterns. It also finds that FedProx aggregation consistently beats FedAvg as the number of clients grows, and introduces an "efficacy energy" metric to score a system's average performance across client configurations. The significance is practical: if these results hold, health authorities could collaborate on outbreak forecasting while keeping raw individual-level data local.

What carries the argument

The central mechanism is a federated training loop in which each client owns a subnetwork induced by a partition of the full network, trains locally on its own node-state trajectories, and shares only model parameters with a central server that aggregates them via FedAvg or FedProx. The local predictors are two architectures: a pure temporal LSTM and the proposed Spatio-Temporal Graph Attention Network (STGAT), which combines an embedding layer, a multi-head graph attention layer for spatial dependencies, two LSTM layers for temporal dependencies, and a softmax classifier. The paper's novel evaluation tool is "efficacy energy," defined as the average of the mean client metric over an increasing number of clients, which measures system robustness under uncertain client configurations rather than at one fixed setting. Simulations use the Gillespie algorithm on a practical airline network, and graph partitions include even node-index splitting, spectral clustering, and Kernighan-Lin partitioning.

What would settle it

Run the same federated pipeline on the same airline network and epidemic parameters, but partition nodes by geographic region or by a cut that removes the busiest inter-region routes, holding everything else fixed. If STGAT no longer outperforms LSTM, or if FedProx's advantage shrinks, then the reported results are an artifact of the chosen partition rather than a general property of federated epidemic prediction.

Watch

Extended reading notes

Core claim

On a 600-node OpenFlights airline network with epidemic trajectories generated by a Gillespie simulator, the paper claims that federated learning provides effective node-level epidemic prediction across seven compartmental models. In centralized baselines, LSTM performs better for SIS, SIR, SEIR, and SIR VS, while STGAT performs better for non-Markovian SIS, SIS with time-varying rates, and SIRS. In federated settings, FedProx consistently outperforms FedAvg, and its advantage grows with the number of clients, which the authors attribute to the proximal term stabilizing local updates under data heterogeneity. The paper also shows that graph partition strategy matters: spectral clustering helps when clients are few, Kernighan-Lin partitioning stays strong across client counts, and even index-based splits work reasonably. Performance degrades as the effective infection rate rises and as missing infectious reports increase, with the authors framing this as a tension between information richness and intrinsic stochasticity.

Load-bearing premise

The central premise is that cutting the network into per-client subnetworks and discarding cross-client edges preserves enough spatial structure for the graph-based model to remain meaningful; if transmission across the cut edges dominates, the paper's STGAT results may not transfer to real deployments.

Editorial extensions

If this is right

  • A central server can train node-level epidemic predictors across partitioned subnetworks without centralizing raw health data, enabling inter-region collaboration under privacy constraints.
  • FedProx should be preferred over FedAvg as the default aggregation method for federated epidemic prediction, because it degrades more gracefully as the number of clients increases.
  • For epidemic processes with fluctuating or periodic dynamics, the graph-aware STGAT model earns its added complexity, while for simpler outbreak patterns a temporal LSTM is sufficient.
  • Partition strategy is a first-order design choice: Kernighan-Lin partitioning balances feature consistency and volume uniformity across clients, while spectral clustering helps only when client counts are small.
  • Missing or noisy infectious reports degrade federated prediction performance, so deployments need data-quality safeguards such as noise-resilient aggregation or imputation.

Reading between the lines

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

  • If the partition sensitivity seen in the paper is real, then real deployments where authorities hold only their own region's data may lose STGAT's advantage unless cross-client boundary information is shared through some privacy-preserving mechanism.
  • The efficacy energy metric is a reusable statistic for comparing any federated system under uncertain client participation, not just epidemic prediction, since it averages performance across client counts instead of relying on one configuration.
  • A direct testable extension would be to add synthetic cross-client transmission, such as shared boundary nodes or shared edge statistics, to measure how much of STGAT's gain comes from within-client topology versus global network structure.
  • The framework's logic could extend to other collective dynamics on networks, such as traffic congestion, cascading failures, or information spread, but the paper's static-network assumption leaves dynamic contact networks as an open challenge.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 6 minor

Summary. The paper proposes a federated learning framework for node-level epidemic prediction on networks, where the network is partitioned among clients that keep local data private and a central server aggregates model updates using FedAvg or FedProx. Two predictors are compared: a pure temporal LSTM and a spatio-temporal graph attention network (STGAT). The evaluation on a 600-node OpenFlights airline network covers seven epidemic models (SIS, SIR, SEIR, SIRS, SIR-VS, non-Markovian SIS, SIS with time-varying rates), varying numbers of clients, three graph partitioning schemes, different effective infection rates, and missing reporting. A summary statistic called 'efficacy energy' is introduced to aggregate performance across client counts. The central claims are that no single model dominates across all scenarios, FedProx generally outperforms FedAvg, and STGAT is preferable for fluctuating dynamics while LSTM suffices for simpler patterns.

Significance. If the results hold, the paper is a useful contribution to privacy-preserving epidemic forecasting: it formulates a realistic cross-silo setting, applies existing FL algorithms to a new domain, and proposes a compact metric (efficacy energy) for comparing FL configurations under variable client counts. The study explicitly acknowledges that STGAT does not universally beat LSTM, which is a more nuanced conclusion than typical 'graph-based model wins' narratives. The paper also ships source code on GitHub, which is helpful for reproducibility. However, the current experimental support is thin: a single stochastic realization, no error bars, absent epidemic parameter values, and a model comparison that confounds architecture capacity with graph usage all limit the reliability of the qualitative takeaways.

major comments (4)
  1. [4.1.2–4.1.3] The conclusion that STGAT is superior for fluctuating dynamics (nmSIS, SIStv, SIRS) is not supported by the comparisons as run: the LSTM baseline is a single 64-unit LSTM layer, whereas STGAT adds an 8-head graph attention layer followed by two LSTM layers with 32 and 64 hidden units. The models therefore differ in parameter count and temporal depth, not only in graph usage. Without a capacity-matched LSTM or an ablation on the GAT block, the observed gains cannot be attributed to the graph-based attention mechanism, which is the load-bearing interpretation in Section 4.1.3 and Section 6.
  2. [4.2.1] The evaluation partitions the network by node index and trains each client on its induced subgraph, discarding all cross-client edges. When transmission across partition boundaries is significant, the local subgraph omits the very spatial interactions needed to predict boundary nodes, so the reported federated performance may be an artifact of the partition rather than a property of the FL framework. The paper should report the fraction of cut edges for each M and ideally include a condition in which the server shares topology (adjacency) with clients while keeping node states private, to separate privacy-preserving aggregation from the loss of global structural information.
  3. [4.2.2, Eq. (20)] Equation (20) defines the efficacy energy as η = 1/(M0-2) Σ_{K=2}^{M0} \bar{α}[M], but the summation index K does not appear in the summand, and the denominator M0-2 does not match the M0-1 terms in the sum (from 2 to M0). This makes the metric not well-defined as written; it should be \bar{α}[K] (or \bar{α}[M]) with denominator M0-1. Since the paper's comparative conclusions are drawn from this aggregated quantity, this must be corrected.
  4. [4.1.1–4.1.2, Table 1] No values are reported for any of the epidemic parameters listed in Table 1 (β, δ, ω, v1, v2, a, b, c, and the Weibull parameters for non-Markovian SIS), and the figures of prevalence (Figure 4) are not sufficient to determine them. In addition, the experiments use a single Gillespie realization per configuration with no error bars or multiple seeds. As a result, the claims that FedProx 'consistently outperforms' FedAvg and that STGAT outperforms LSTM in specific regimes are not supported by any measure of statistical significance or reproducibility.
minor comments (6)
  1. [Figure 1 caption] The caption contains a typo: 'the greed for the recovered' should be 'the green for the recovered'.
  2. [Section 4.1.2] The text says 'they are used to forecast traffic speed in the next tF = 10 time steps'; this is a copy-paste error from a traffic-forecasting paper and should read 'forecast epidemic node states'.
  3. [Algorithm 1] The phrase 'if Early Stop condition satisfried' should be 'satisfied'.
  4. [Figure 11] The caption and axis labels use 'missing radio' instead of 'missing ratio'.
  5. [Section 5.2] The phrase 'allowing local users to access high-equality services' should be 'high-quality services'.
  6. [Section 3.1] Definition 2 allows 'disjoint or overlapping subnetworks', but the experiments only implement disjoint partitions; the paper should clarify that overlapping regions are out of scope for the numerical study.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the federated-learning comparisons and the efficacy-energy ranking are empirical measurements, not derivations that reduce to their own inputs.

full rationale

The paper's central claims—that FedProx outperforms FedAvg and that STGAT and LSTM have complementary strengths—are supported by measured performance on simulated epidemic data. The efficacy energy defined in Eq. (20) is a descriptive aggregation, namely the mean over client counts of the measured average metric ᾱ[M], and the statement that FedProx achieves higher energy is an empirical observation about the measured performance curves, not a result forced by the definition. No parameter is fitted to a target and then renamed as a prediction. The self-citations in the paper are not load-bearing: reference [24] is used for the Gillespie-style simulation approach, reference [6] is background on SIS control, and reference [61] appears only as a suggested future direction. No uniqueness theorem from the authors' prior work is invoked to forbid alternatives, and the STGAT architecture is presented as a designed model with external GAT/LSTM ancestry rather than as an ansatz justified solely by self-citation. The concern that the LSTM/STGAT comparison is confounded by model capacity is a correctness and experimental-design issue, not a circularity issue, because the comparison does not reduce by construction to its inputs. Overall, the derivation chain is self-contained with respect to the empirical evaluations, so no circular step is identified.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central claim rests on the simulation setup and the chosen evaluation regime rather than on fitted theoretical parameters. The most load-bearing free parameters are the unreported epidemic rates, the chosen upper client count M0 for the efficacy energy, and the FedProx proximal term. The assumptions listed are the domain modeling choices (Gillespie simulation, airport network as contact structure, exclusion of extinction phases, induced-subgraph clients) that define the scope of the claim.

free parameters (3)
  • efficacy energy upper client count M0 = 16 or 25 (by scenario)
    The efficacy energy eta averages performance from K=2 to M0; the chosen M0 changes the reported energy and can affect model rankings across scenarios (Section 4.2.2, Figures 7 and 9).
  • epidemic simulation parameters (beta, delta, omega, v1, v2, a, b, c) = not reported
    Each compartmental model in Table 1 needs numeric infection, curing, waning, and vaccination rates; the paper never lists them, so the simulations are not reproducible from the text and the qualitative rankings may depend on the chosen regime.
  • FedProx proximal term mu = 0.01
    Default value set for all experiments; the conclusion that FedProx helps relies on this choice (Section 4.2).
assumptions (5)
  • domain assumption Gillespie event-driven simulation accurately realizes the stated Markovian and non-Markovian epidemic processes
    Used to generate all training data (Section 4.1.1); if the simulator mis-implements a process (e.g., the non-Markovian Weibull timing), the data would not match the intended model.
  • domain assumption Node-level classification accuracy meaningfully measures prediction quality, despite most nodes being static at any time step
    The paper itself concedes in Section 4.1.3 that accuracy is inflated by static nodes and switches to 1/CE and RMSE; the inflated accuracy is still reported in Figure 5 and Table 2.
  • ad hoc to paper The dynamic (pre-extinction) phase is the regime of interest; phases near extinction are excluded from evaluation
    Section 4.1.1: 'we focus only the dynamic stage of the spreading process'; this post-hoc selection narrows the scope of the claim about general FL efficacy.
  • domain assumption The OpenFlights airport network with 600 highest-degree airports and 22,352 links is a representative contact structure for epidemic spread
    Section 4.1.1; the realism of the privacy scenario depends on this proxy.
  • ad hoc to paper Each client's local graph is the induced subgraph of the partition, with cross-client edges removed, and this preserves the spatial signal needed by STGAT
    Section 4.2.1 describes even partitioning by node index; the effect of discarding cross-client edges on the STGAT advantage is never quantified.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards the efficacy of federated prediction for epidemics on networks." pith.science (2026). https://pith.science/paper/CXVNP6TG

@misc{pith2026241202161,
  author       = {Pith},
  title        = {Pith review of: Towards the efficacy of federated prediction for epidemics on networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CXVNP6TG}},
  note         = {Machine review of arXiv:2412.02161}
}
read the original abstract

Epidemic prediction is of practical significance in public health, enabling early intervention, resource allocation, and strategic planning. However, privacy concerns often hinder the sharing of health data among institutions, limiting the development of accurate prediction models. In this paper, we develop a general privacy-preserving framework for node-level epidemic prediction on networks based on federated learning (FL). We frame the spatio-temporal spread of epidemics across multiple data-isolated subnetworks, where each node state represents the aggregate epidemic severity within a community. Then, both the pure temporal LSTM model and the spatio-temporal model i.e., Spatio-Temporal Graph Attention Network (STGAT) are proposed to address the federated epidemic prediction. Extensive experiments are conducted on various epidemic processes using a practical airline network, offering a comprehensive assessment of FL efficacy under diverse scenarios. By introducing the efficacy energy metric to measure system robustness under various client configurations, we systematically explore key factors influencing FL performance, including client numbers, aggregation strategies, graph partitioning, missing infectious reports. Numerical results manifest that STGAT excels in capturing spatio-temporal dependencies in dynamic processes whereas LSTM performs well in simpler pattern. Moreover, our findings highlight the importance of balancing feature consistency and volume uniformity among clients, as well as the prediction dilemma between information richness and intrinsic stochasticity of dynamic processes. This study offers practical insights into the efficacy of FL scenario in epidemic management, demonstrates the potential of FL to address broader collective dynamics.

Figures

Figures reproduced from arXiv: 2412.02161 by the authors.

Figure 1
Figure 1. Illustration of an SIR spreading on a network. The left subfigure shows a snapshot at [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Illustration of the proposed federated learning framework designed to epidemic prediction. A central server coordinates [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Architecture of spatio-temporal graph attention networks (STGAT). [PITH_FULL_IMAGE:figures/full_fig_p010_3.png] view at source ↗
Figures from the paper (8 more)
Figure 4
Figure 4. Figure 4: Illustrations of the fraction of infected nodes as a function of time for various epidemic processes, including SIS, SIR, [PITH_FULL_IMAGE:figures/full_fig_p014_4.png]
Figure 5
Figure 5. Figure 5: Comparison of centralized learning and federated learning for an exampled nmSIS process under 4 or 8 clients by the [PITH_FULL_IMAGE:figures/full_fig_p016_5.png]
Figure 6
Figure 6. Figure 6: Violin plots of the metrics, e.g. 1/CE and RMSE, as a function of the number of clients M in a nmSIS federated learning. The histograms represent the metrics among all clients under different scenarios, and the line represent the mean of the metric ¯α. The gray surface…
Figure 7
Figure 7. Figure 7: Performance energy η(α = 1/CE) for 7 epidemic process under different prediction models and aggregation approaches, with the upper-limit number of clients M0 = 16. of leveraging graph-based topological information to effectively capture the complex spatio-temporal patt…
Figure 8
Figure 8. Figure 8: Performance metric ¯α = E[1/CE] as a function of the number of clients M under different graph partition G for three epidemic cases, e.g., SIS, nmSIS and SEIR. LSTM+Even LSTM+SP LSTM+KL STGAT+Even STGAT+SP STGAT+KL 0 5 10 15 20 25 30 35 40 E n e r g y ( = 1 / C E) SIS …
Figure 9
Figure 9. Figure 9: Efficacy energy η as the metric α = 1/CE of the reciprocal CE for different federated scenarios. The combination of prediction model (the LSTM and the STGAT) and three graph partitioning methods for three typical epidemics are investigated. The upper-limit number of cl…
Figure 10
Figure 10. Figure 10: Performance metrics ¯α = E[1/CE] and ¯α = E[RMSE] as a function of the effective infection rate τ for nmSIS and SIS processes under a federated learning with M = 10 clients. The epidemic threshold τc ≈ 0.015 is marked as a vertical line. Note that a higher metric ¯α =…
Figure 11
Figure 11. Figure 11: Performance metric ¯α of accuracy and RMSE as a function of the node missing radio under a nmSIS federated learning of M = 6 clients. 5 Related work 5.1 Epidemic prediction on networks Epidemic prediction has gained significant attention in recent years due to its cri…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

61 extracted references · 57 canonical work pages

  1. [1]

    Essai d’une nouvelle analyse de la mortalit´ e caus´ ee par la petite v´ erole, et des avantages de l’inoculation pour la pr´ evenir,

    D. Bernoulli, “Essai d’une nouvelle analyse de la mortalit´ e caus´ ee par la petite v´ erole, et des avantages de l’inoculation pour la pr´ evenir,”Histoire de l’Acad., Roy. Sci.(Paris) avec Mem , pp. 1–45, 1760

    work page

  2. [2]

    Epidemic processes in complex networks,

    R. Pastor Satorras, C. Castellano, P. Van Mieghem, and A. Vespignani, “Epidemic processes in complex networks,” Reviews of Modern Physics , vol. 87, no. 3, p. 925, 2015

    work page 2015

  3. [3]

    Mitigation of cascading failures in complex networks,

    A. Smolyak, O. Levy, I. Vodenska, S. Buldyrev, and S. Havlin, “Mitigation of cascading failures in complex networks,” Scientific reports, vol. 10, no. 1, p. 16124, 2020

    work page 2020

  4. [4]

    The spread of true and false news online,

    S. Vosoughi, D. Roy, and S. Aral, “The spread of true and false news online,” science, vol. 359, no. 6380, pp. 1146–1151, 2018

    work page 2018

  5. [5]

    Prediction of covid-19 corona virus pandemic based on time series data using support vector machine,

    V. Singh, R. C. Poonia, S. Kumar, P. Dass, P. Agarwal, V. Bhatnagar, and L. Raja, “Prediction of covid-19 corona virus pandemic based on time series data using support vector machine,” Journal of Discrete Mathematical Sciences and Cryptography , vol. 23, no. 8, pp. 1583–1597, 2020

    work page 2020

  6. [6]

    Optimal induced spreading of SIS epidemics in networks,

    H. Zhidong and P. Van Mieghem, “Optimal induced spreading of SIS epidemics in networks,” IEEE Transactions on Control of Network Systems , vol. 6, no. 4, pp. 1344–1353, 2018

    work page 2018

  7. [7]

    A review on the long short-term memory model,

    G. Van Houdt, C. Mosquera, and G. N´ apoles, “A review on the long short-term memory model,” Artificial Intelligence Review , vol. 53, no. 8, pp. 5929–5955, 2020

    work page 2020

  8. [8]

    A comprehensive survey on graph neural networks,

    Z. Wu, S. Pan, F. Chen, G. Long, C. Zhang, and S. Y. Philip, “A comprehensive survey on graph neural networks,” IEEE transactions on neural networks and learning systems , vol. 32, no. 1, pp. 4–24, 2020

    work page 2020

Show all 61 references
  1. [9]

    Tgcn: Tag graph convo- lutional network for tag-aware recommendation,

    B. Chen, W. Guo, R. Tang, X. Xin, Y. Ding, X. He, and D. Wang, “Tgcn: Tag graph convo- lutional network for tag-aware recommendation,” in Proceedings of the 29th ACM International Conference on Information & Knowledge Management , 2020, pp. 155–164

    2020

  2. [10]

    Spatial-temporal fusion graph neural networks for traffic flow forecasting,

    M. Li and Z. Zhu, “Spatial-temporal fusion graph neural networks for traffic flow forecasting,” in Proceedings of the AAAI conference on artificial intelligence , vol. 35, no. 5, 2021, pp. 4189–4196

    2021

  3. [11]

    Drug discovery with explainable artificial intel- ligence,

    J. Jim´ enez-Luna, F. Grisoni, and G. Schneider, “Drug discovery with explainable artificial intel- ligence,” Nature Machine Intelligence , vol. 2, no. 10, pp. 573–584, 2020

    2020

  4. [12]

    Federated optimization: Distributed machine learning for on-device intelligence,

    J. Koneˇ cn` y, H. B. McMahan, D. Ramage, and P. Richt´ arik, “Federated optimization: Distributed machine learning for on-device intelligence,” arXiv preprint arXiv:1610.02527 , 2016

    2016 arXiv

  5. [13]

    Federated learning-based private medical knowledge graph for epidemic surveillance in internet of things,

    X. Wu, J. Gao, M. Bilal, F. Dai, X. Xu, L. Qi, and W. Dou, “Federated learning-based private medical knowledge graph for epidemic surveillance in internet of things,” Expert Systems , p. e13372, 2023. 23

    2023

  6. [14]

    Iomt: A covid-19 healthcare system driven by federated learning and blockchain,

    O. Samuel, A. B. Omojo, A. M. Onuja, Y. Sunday, P. Tiwari, D. Gupta, G. Hafeez, A. S. Yahaya, O. J. Fatoba, and S. Shamshirband, “Iomt: A covid-19 healthcare system driven by federated learning and blockchain,” IEEE Journal of Biomedical and Health Informatics , vol. 27, no. 2...

    2022

  7. [15]

    Privacy-preserving individual-level covid- 19 infection prediction via federated graph learning,

    W. Fu, H. Wang, C. Gao, G. Liu, Y. Li, and T. Jiang, “Privacy-preserving individual-level covid- 19 infection prediction via federated graph learning,” ACM Transactions on Information Systems, vol. 42, no. 3, pp. 1–29, 2024

    2024

  8. [16]

    Analysis of mobility based covid-19 epidemic model using federated multitask learning,

    M. Kumaresan, M. S. Kumar, and N. Muthukumar, “Analysis of mobility based covid-19 epidemic model using federated multitask learning,” Math. Biosci. Eng , vol. 19, pp. 9983–10 005, 2022

    2022

  9. [17]

    Stgat: Modeling spatial-temporal interactions for human trajectory prediction,

    Y. Huang, H. Bi, Z. Li, T. Mao, and Z. Wang, “Stgat: Modeling spatial-temporal interactions for human trajectory prediction,” in Proceedings of the IEEE/CVF international conference on computer vision, 2019, pp. 6272–6281

    2019

  10. [18]

    M. E. Newman, A.-L. E. Barab´ asi, and D. J. Watts, The structure and dynamics of networks. Princeton University Press, 2006

    2006

  11. [19]

    Epidemic dynamics and endemic states in complex net- works,

    R. Pastor-Satorras and A. Vespignani, “Epidemic dynamics and endemic states in complex net- works,” Physical Review E , vol. 63, no. 6, p. 066117, 2001

    2001

  12. [20]

    Van Mieghem, Performance analysis of complex networks and systems

    P. Van Mieghem, Performance analysis of complex networks and systems . Cambridge University Press, 2014

    2014

  13. [21]

    Thresholds for epidemic spreading in networks,

    C. Castellano and R. Pastor-Satorras, “Thresholds for epidemic spreading in networks,” Physical Review Letters, vol. 105, no. 21, p. 218701, 2010

    2010

  14. [22]

    Non-Markovian infection spread dramatically alters the susceptible-infected-susceptible epidemic threshold in networks,

    P. Van Mieghem and R. Van de Bovenkamp, “Non-Markovian infection spread dramatically alters the susceptible-infected-susceptible epidemic threshold in networks,” Physical Review Letters, vol. 110, no. 10, p. 108701, 2013

    2013

  15. [23]

    Virus spread in networks,

    P. Van Mieghem, J. Omic, and R. Kooij, “Virus spread in networks,” IEEE/ACM Transactions on Networking , vol. 17, no. 1, pp. 1–14, 2009

    2009

  16. [24]

    The spreading time in SIS epidemics on networks,

    Z. He and P. Van Mieghem, “The spreading time in SIS epidemics on networks,” Physica A: Statistical Mechanics and its Applications , vol. 494, pp. 317–330, 2018

    2018

  17. [25]

    An SIR VS epidemic model with pulse vaccination strategy,

    T. Zhang and Z. Teng, “An SIR VS epidemic model with pulse vaccination strategy,” Journal of theoretical biology, vol. 250, no. 2, pp. 375–381, 2008

    2008

  18. [26]

    Critical behavior of the sis epidemic model with time- dependent infection rate,

    N. Crokidakis and M. A. De Menezes, “Critical behavior of the sis epidemic model with time- dependent infection rate,” Journal of Statistical Mechanics: Theory and Experiment , vol. 2012, no. 05, p. P05012, 2012

    2012

  19. [27]

    Group-based general epidemic modeling for spreading processes on networks: Groupgem,

    S. A. Moon, F. D. Sahneh, and C. Scoglio, “Group-based general epidemic modeling for spreading processes on networks: Groupgem,” IEEE Transactions on Network Science and Engineering , vol. 8, no. 1, pp. 434–446, 2020. 24

    2020

  20. [28]

    Epidemic outbreaks in networks with equitable or almost-equitable partitions,

    S. Bonaccorsi, S. Ottaviano, D. Mugnolo, and F. D. Pellegrini, “Epidemic outbreaks in networks with equitable or almost-equitable partitions,” SIAM Journal on Applied Mathematics , vol. 75, no. 6, pp. 2421–2443, 2015

    2015

  21. [29]

    Epidemic diffusion network of spain: A mobility model to characterize the transmission routes of disease,

    J. Del- ´Aguila-Mej ´ ıa, D. Garc ´ ıa-Garc ´ ıa, A. Rojas-Benedicto, N. Rosillo, M. Guerrero-Vadillo, M. Pe˜ nuelas, R. Ramis, D. G´ omez-Barroso, and J. d. M. Donado-Campos, “Epidemic diffusion network of spain: A mobility model to characterize the transmission routes of dis...

    2023

  22. [30]

    Communication-efficient learning of deep networks from decentralized data,

    B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, “Communication-efficient learning of deep networks from decentralized data,” in Artificial intelligence and statistics . PMLR, 2017, pp. 1273–1282

    2017

  23. [31]

    Federated optimization in heterogeneous networks,

    T. Li, A. K. Sahu, M. Zaheer, M. Sanjabi, A. Talwalkar, and V. Smith, “Federated optimization in heterogeneous networks,” Proceedings of Machine learning and systems , vol. 2, pp. 429–450, 2020

    2020

  24. [32]

    Human mobility and the global spread of infectious diseases: a focus on air travel,

    A. Findlater and I. I. Bogoch, “Human mobility and the global spread of infectious diseases: a focus on air travel,” Trends in parasitology, vol. 34, no. 9, pp. 772–783, 2018

    2018

  25. [33]

    The influence of passenger air traffic on the spread of covid-19 in the world,

    Y. M. Sokadjo and M. N. Atchad´ e, “The influence of passenger air traffic on the spread of covid-19 in the world,” Transportation Research Interdisciplinary Perspectives, vol. 8, p. 100213, 2020

    2020

  26. [34]

    Algebraic connectivity of graphs,

    M. Fiedler, “Algebraic connectivity of graphs,” Czechoslovak Mathematical Journal , vol. 23, pp. 298–305, 1973

    1973

  27. [35]

    An efficient heuristic procedure for partitioning graphs,

    B. W. Kernighan and S. Lin, “An efficient heuristic procedure for partitioning graphs,” Bell System Technical Journal, vol. 49, pp. 291–307, 1970

    1970

  28. [36]

    Localization and spreading of diseases in complex networks,

    A. V. Goltsev, S. N. Dorogovtsev, J. Oliveira, and J. F. Mendes, “Localization and spreading of diseases in complex networks,” Physical Review Letters, vol. 109, no. 12, p. 128702, 2012

    2012

  29. [37]

    Analysis, prediction, and control of epidemics: A survey from scalar to dynamic network models,

    L. Zino and M. Cao, “Analysis, prediction, and control of epidemics: A survey from scalar to dynamic network models,” IEEE Circuits and Systems Magazine , vol. 21, no. 4, pp. 4–23, 2021

    2021

  30. [38]

    A systematic literature review of time series methods applied to epidemic prediction,

    A. B. Bamana, M. S. Kamalabad, and D. L. Oberski, “A systematic literature review of time series methods applied to epidemic prediction,” Informatics in Medicine Unlocked , p. 101571, 2024

    2024

  31. [39]

    Epidemic prediction algorithm based on deep learning and epi- demic dynamics,

    B. Xu, X. Chen, and Z. Chen, “Epidemic prediction algorithm based on deep learning and epi- demic dynamics,” in Proceedings of the 2023 4th International Conference on Machine Learning and Computer Application , 2023, pp. 1021–1026

    2023

  32. [40]

    Predictions for covid-19 with deep learning models of lstm, gru and bi-lstm,

    F. Shahid, A. Zameer, and M. Muneeb, “Predictions for covid-19 with deep learning models of lstm, gru and bi-lstm,” Chaos, Solitons & Fractals , vol. 140, p. 110212, 2020

    2020

  33. [41]

    Interaction-temporal gcn: a hybrid deep framework for covid-19 pandemic analysis,

    Z. Yu, X. Zheng, Z. Yang, B. Lu, X. Li, and M. Fu, “Interaction-temporal gcn: a hybrid deep framework for covid-19 pandemic analysis,” IEEE Open Journal of Engineering in Medicine and Biology, vol. 2, pp. 97–103, 2021. 25

    2021

  34. [42]

    Stan: spatio-temporal attention network for pandemic prediction using real-world evidence,

    J. Gao, R. Sharma, C. Qian, L. M. Glass, J. Spaeder, J. Romberg, J. Sun, and C. Xiao, “Stan: spatio-temporal attention network for pandemic prediction using real-world evidence,” Journal of the American Medical Informatics Association , vol. 28, no. 4, pp. 733–743, 2021

    2021

  35. [43]

    A review of graph neural networks in epidemic modeling,

    Z. Liu, G. Wan, B. A. Prakash, M. S. Lau, and W. Jin, “A review of graph neural networks in epidemic modeling,” in Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining , 2024, pp. 6577–6587

    2024

  36. [44]

    Why is it difficult to accurately predict the covid-19 epidemic?

    W. C. Roda, M. B. Varughese, D. Han, and M. Y. Li, “Why is it difficult to accurately predict the covid-19 epidemic?” Infectious disease modelling , vol. 5, pp. 271–281, 2020

    2020

  37. [45]

    Real-time epidemic forecasting: challenges and opportunities,

    A. N. Desai, M. U. Kraemer, S. Bhatia, A. Cori, P. Nouvellet, M. Herringer, E. L. Cohn, M. Car- rion, J. S. Brownstein, L. C. Madoff et al. , “Real-time epidemic forecasting: challenges and opportunities,” Health security, vol. 17, no. 4, pp. 268–275, 2019

    2019

  38. [46]

    On the uncertainty of real-time predictions of epidemic growths: A covid-19 case study for china and italy,

    T. Alberti and D. Faranda, “On the uncertainty of real-time predictions of epidemic growths: A covid-19 case study for china and italy,” Communications in Nonlinear Science and Numerical Simulation, vol. 90, p. 105372, 2020

    2020

  39. [47]

    Big data analytics for epidemic forecasting: Policy frameworks and technical approaches,

    G. T. Igwama, J. A. Olaboye, C. C. Maha, M. D. Ajegbile, and S. Abdul, “Big data analytics for epidemic forecasting: Policy frameworks and technical approaches,” International Journal of Applied Research in Social Sciences , vol. 6, no. 7, pp. 1449–1460, 2024

    2024

  40. [48]

    A review of applications in federated learning,

    L. Li, Y. Fan, M. Tse, and K.-Y. Lin, “A review of applications in federated learning,” Computers & Industrial Engineering , vol. 149, p. 106854, 2020

    2020

  41. [49]

    Privacy preserving time-series forecasting of user health data streams,

    S. Imtiaz, S.-F. Horchidan, Z. Abbas, M. Arsalan, H. N. Chaudhry, and V. Vlassov, “Privacy preserving time-series forecasting of user health data streams,” in 2020 IEEE International Con- ference on Big Data (Big Data) . IEEE, 2020, pp. 3428–3437

    2020

  42. [50]

    Leveraging asynchronous federated learning to predict customers financial distress,

    A. Imteaj and M. H. Amini, “Leveraging asynchronous federated learning to predict customers financial distress,” Intelligent Systems with Applications , vol. 14, p. 200064, 2022

    2022

  43. [51]

    Deep anomaly detection for time-series data in industrial iot: A communication-efficient on-device federated learning approach,

    Y. Liu, S. Garg, J. Nie, Y. Zhang, Z. Xiong, J. Kang, and M. S. Hossain, “Deep anomaly detection for time-series data in industrial iot: A communication-efficient on-device federated learning approach,” IEEE Internet of Things Journal , vol. 8, no. 8, pp. 6348–6358, 2020

    2020

  44. [52]

    Stochastic controlled averaging for federated learning with commu- nication compression,

    X. Huang, P. Li, and X. Li, “Stochastic controlled averaging for federated learning with commu- nication compression,” arXiv preprint arXiv:2308.08165 , 2023

    2023 arXiv

  45. [53]

    Retracted: Personalized federated learning framework for network traffic anomaly detection,

    J. Pei, K. Zhong, M. A. Jan, and J. Li, “Retracted: Personalized federated learning framework for network traffic anomaly detection,” 2022

    2022

  46. [54]

    Multi-task federated learning-based system anomaly detection and multi-classification for microservices architecture,

    J. Hao, P. Chen, J. Chen, and X. Li, “Multi-task federated learning-based system anomaly detection and multi-classification for microservices architecture,” Future Generation Computer Systems, vol. 159, pp. 77–90, 2024. 26

    2024

  47. [55]

    Unveiling group-specific distributed concept drift: A fairness imperative in federated learning,

    T. Salazar, J. Gama, H. Ara´ ujo, and P. H. Abreu, “Unveiling group-specific distributed concept drift: A fairness imperative in federated learning,” arXiv preprint arXiv:2402.07586 , 2024

    2024 arXiv

  48. [56]

    Flame: Adaptive and reactive concept drift mitigation for federated learning deployments,

    I. Mavromatis, S. De Feo, and A. Khan, “Flame: Adaptive and reactive concept drift mitigation for federated learning deployments,” arXiv preprint arXiv:2410.01386 , 2024

    2024 arXiv

  49. [57]

    Federated learning for heterogeneous electronic health records utilising augmented temporal graph attention networks,

    S. Molaei, A. Thakur, G. Niknam, A. Soltan, H. Zare, and D. A. Clifton, “Federated learning for heterogeneous electronic health records utilising augmented temporal graph attention networks,” in International Conference on Artificial Intelligence and Statistics. PMLR, 2024, pp...

    2024

  50. [58]

    A decentralized federated learning-based spatial–temporal model for freight traffic speed forecasting,

    X. Shen, J. Chen, S. Zhu, and R. Yan, “A decentralized federated learning-based spatial–temporal model for freight traffic speed forecasting,” Expert Systems with Applications, vol. 238, p. 122302, 2024

    2024

  51. [59]

    The merit of river network topology for neural flood forecasting,

    N. Kirschstein and Y. Sun, “The merit of river network topology for neural flood forecasting,” arXiv preprint arXiv:2405.19836 , 2024

    2024 arXiv

  52. [60]

    Universal resilience patterns in complex networks,

    J. Gao, B. Barzel, and A.-L. Barab´ asi, “Universal resilience patterns in complex networks,” Nature, vol. 530, no. 7590, p. 307, 2016

    2016

  53. [61]

    Heterogeneity induced localization of traffic congestion on networks,

    Z. He, “Heterogeneity induced localization of traffic congestion on networks,” arXiv preprint arXiv:2303.12323, 2023. 27

    2023 arXiv

Pith tools

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