Reliable Vertical Federated Learning in 5G Core Network Architecture

Mohamad Mestoukirdi; Mourad Khanfouci

REVIEW 4 major objections 4 minor 17 references

Reliable Vertical Federated Learning in 5G Core Network Architecture

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 The paper claims that reliability-aware feature allocation reduces vertical federated learning loss in 5G core networks by assigning important features and larger embedding dimensions to more dependable clients.

desk verdict A sensible reliability-aware feature allocation heuristic for VFL in 5G core networks, undermined by test-set checkpoint selection and missing uncertainty quantification; fixable with a validation split. read the letter →

arxiv 2505.15244 v3 pith:GETJJJXW submitted 2025-05-21 cs.LG cs.SYeess.SY

Mohamad Mestoukirdi , Mourad Khanfouci This is my paper · ORCID

classification cs.LGcs.SYeess.SY

keywords verticalfederatedlearningclientreliabilityfeatureimportanceassociationmodeldefinition5GcorenetworkNWDAFgeneralizationloss

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

Vertical federated learning (VFL) lets several data holders train a shared model on different feature sets, but in a 5G core network the analytics functions acting as clients can drop out under load, and the missing embeddings degrade the model. This paper proposes that the central server should assign features and local-model sizes by client reliability: more important features and larger embedding dimensions go to clients with higher forecast reliability. In simulations of a video-quality prediction task with four clients, success probabilities drawn from Beta(8,2) (mean 0.8) and Beta(10,6) (mean 0.625), the proposed method reduces average weighted test loss by about 14.5% and 18% respectively relative to a dropout-simulation baseline. If the result holds, reliability-aware feature allocation is a low-cost way to make VFL robust in centralized network environments, without changing the federated training loop.

What carries the argument

The mechanism is a feature association and model definition algorithm executed before training. The server first has an analytics function compute feature-importance scores for the centrally stored data, then assigns features to each client so that total assigned feature importance and embedding dimension both scale linearly with that client's reliability probability, subject to a fixed total embedding budget. The server model's input dimension is set by the concatenation of these per-client embedding sizes, and dropouts in the training loop cause unavailable clients' embeddings to be replaced by zero vectors. This couples the information capacity of each client's contribution to the probability that the contribution arrives.

What would settle it

Re-run the same simulations with reliability forecasts corrupted by noise or drifting mid-training; if the average weighted test loss improvement over the baseline shrinks or reverses once forecast error passes a small threshold, the central claim collapses. A cleaner separation test is to compare against a baseline that keeps the same per-client embedding dimensions but assigns features randomly, which would isolate whether the gain comes from feature-importance matching or embedding sizing.

Watch

Extended reading notes

Core claim

In a VFL system with clients of unequal reliability, the way features are partitioned across clients is itself a source of robustness. The paper's central claim is that a server with centralized control over a feature pool can mitigate dropout-induced generalization loss by assigning each client a local model whose feature set and embedding dimension are scaled to that client's forecast reliability. In the evaluated 5G core-network scenario, this reliability-aware assignment yields an average weighted test loss about 14.5% lower than the dropout-simulation baseline under Beta(8,2) reliability and about 18% lower under Beta(10,6) reliability. The improvement concentrates in availability patterns where the most reliable client, carrying the most important features, participates in inference.

Load-bearing premise

The load-bearing premise is that each client can accurately forecast its reliability for the whole training period and that these probabilities stay fixed during training and testing, because noisy or drifting forecasts would misorder feature assignment and could reverse the reported gains.

Editorial extensions

If this is right

Operators can adopt the initialization phase as a lightweight add-on to existing NWDAF-based VFL, since the training and inference loops are unchanged.
The method is most valuable when client reliability is moderate and varied; in the high-reliability scenario the gain is smaller but still about 14.5%.
A fixed total embedding budget can be redistributed toward reliable clients without extra communication cost, effectively buying robustness without spending more bandwidth.
The gains depend on centralized data orchestration, so the algorithm transfers only to environments where a server can reassign features among clients.

Reading between the lines

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

The paper does not vary forecast error, so a fair next experiment would corrupt the reported reliability measures and map how much misordering the algorithm tolerates before it falls behind the baseline.
A natural extension is to re-estimate reliability online and periodically reassign features or adjust embedding sizes, which the static initialization here does not allow.
If reliability forecasts are self-reported, clients have an incentive to overstate reliability to receive important features; an incentive-compatible or verified-reporting mechanism would be needed in adversarial deployments.
The same reliability-proportional capacity allocation could apply within a single client's model, for example by giving more network capacity to features that come from dependable data sources.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

Two things to know. First, the mechanism is genuinely new in this setting: the server centrally assigns features by importance and scales each client's local model and embedding size by the client's reliability forecast, exploiting centralized data management in 5G core networks. It is a sensible engineering heuristic and a reasonable step beyond the dropout-simulation and serverless baselines in [10] and [9]. Second, the reported 14.5% and 18% loss reductions are not yet believable as expected deployed gains, because the checkpoints are chosen using test-set performance (Section VI.B) and no error bars or significance tests are given. The stress-test note lands: this is a real evaluation leak.

The paper does several things well. The system model is clearly written, the algorithm is simple to implement, and the code and dataset are released. The authors are also honest that gains concentrate in availability patterns where reliable clients participate (Fig. 4, ID >= 8). For a practitioner in 5G core networks, the idea is plausible and cheap to try.

The soft spots, in proportion. First and main: Section VI.B says the server continually stores models with the best test performance during training and reverts to those top performers. That makes the headline numbers best-of-run over the test set, not the expected performance of a deployed model. The same protocol is used for the baseline, so it is not a relative unfairness, but it is a biased estimate of absolute gain, and if the proposed method's per-round trajectory is noisier—plausible, since capacity is concentrated in a few clients—best-of-run selection can systematically favor it. The fix is straightforward: use a validation split for checkpoint selection and evaluate once on test. The 10-run averages should be accompanied by standard errors or confidence intervals; per-ID differences already flip sign (e.g., Fig. 4 at IDs 2–6), so the average gain needs an interval.

Second, the method assumes each NWDAF can forecast its reliability for the whole training period and that these probabilities stay fixed (Sections V-A and VI-B). The paper does not test sensitivity to forecast error or drift. This is a separate load-bearing premise: if reliability tags are misordered, the allocation could underperform a random split. A simple perturbation experiment would address it.

Minor: one baseline and one dataset (proprietary, though the GitHub release mitigates this), and feature importance is computed with a basic decision tree before training; no analysis of that choice.

Bottom line: a plausible engineering contribution, clearly written and reproducible, but the quantitative claim needs re-evaluation with a validation split and uncertainty quantification. The paper is for researchers working on VFL robustness in 5G core networks, or anyone using feature allocation under reliability constraints. I would send it to review: the mechanism is well defined, the flaw is fixable, and the contribution is non-trivial. But I would not take the 14.5%/18% numbers at face value until the evaluation is corrected.

Recommendation: major revision requiring a held-out validation split, confidence intervals, a sensitivity analysis on reliability forecast error, and ideally an extra baseline.

Referee Report

4 major / 4 minor

Summary. This paper proposes a reliability-aware feature allocation and model definition algorithm for vertical federated learning (VFL) in a 5G core network. The server uses centralized data orchestration to compute feature importances, then assigns more important features and larger embedding dimensions to clients (NWDAFs) with higher predicted reliability, defining local model sizes under a communication budget. The authors evaluate the method against a VFL baseline from [10] on a DASH QoE dataset with K=4 clients under two reliability scenarios (Beta(8,2) and Beta(10,6)), reporting approximately 14.5% and 18% lower average weighted test loss. The manuscript includes pseudocode for initialization, feature association, training, and inference, and a public code repository.

Significance. The motivating setting—VFL in 5G core networks with centralized data management and NWDAF reliability constraints—is timely, and the proposed mechanism is simple and implementable. The strongest assets are the architectural framing tied to 3GPP NWDAFs, the availability of dataset and code, and a concrete falsifiable claim. However, the paper is an empirical heuristic comparison with no theoretical analysis, and the current evaluation protocol undermines the headline numbers: checkpoints are selected on test performance, the reported reduction is computed with an absolute-value metric, and only 10-run averages without error bars are provided. If these issues are fixed, the contribution would be a useful engineering result; as written, the effect size is not yet established.

major comments (4)

[Section VI-B (checkpoint selection)] The evaluation protocol selects the 'best-performing server and client models in terms of its test performance' during training and reverts to these top-performing models for all reported results. This is test-set leakage: the displayed losses are minima over checkpoints evaluated on the same test data, not the expected loss of a model chosen by a held-out validation rule. If the proposed method's per-round trajectory is noisier than the baseline's (plausible, because reliability-scaled embeddings concentrate capacity in a subset of clients), best-of-run selection can systematically favor it even if the final or average model is no better. Please select checkpoints using a validation split, or report the final-round model and the validation-selected model separately, and report per-run results.
[Section VI.C (percentage expression)] The text states that the proposed method achieves 'approximately 14.5% lower test loss' and '~18% reduction', but the displayed formula is Σ_{m=1}^{15} |ℓ_proposed(ID=m) − ℓ_SoTA(ID=m)| / Σ_{m=1}^{15} ℓ_SoTA(ID=m). This is a mean absolute relative deviation, not a signed average reduction. Figs. 2 and 4 show several positive per-ID differences (proposed method worse), so the absolute-value aggregation can overstate the improvement and is inconsistent with the 'lower loss' phrasing. Report the signed average difference (with the distribution across the 10 runs) rather than this absolute-value quantity.
[Section V-A / VI-B (reliability forecasts)] The algorithm relies on each NWDAF providing a reliability forecast that remains fixed throughout training and testing; the simulations draw p_k once and use those exact values as the oracle reliability. The paper does not test sensitivity to forecast error, drift over time, or strategic misreporting. Since the entire feature/embedding allocation is ordered by these forecasts, incorrect forecasts could reorder assignments and reverse the gains. Add experiments with perturbed or time-varying p_k, or clearly state that the claimed benefits hold only under perfectly accurate, static reliability forecasts.
[Section VI.B-C (statistical reporting)] All conclusions rest on averages over 10 simulation runs with no standard errors, confidence intervals, or significance tests, and the comparison uses a single baseline. This is insufficient support for the headline 14.5%/18% effect, particularly in light of the checkpoint-selection issue. Please report per-run distributions, error bars, and a paired test over the 10 runs, and consider at least one additional baseline (e.g., equal embedding dimensions with reliability-aware feature assignment, or random feature assignment with matched total embedding budget).

minor comments (4)

[Section VI] The sentence claiming the baseline 'represents an upper bound on the performance of the algorithm described in [10]' is not justified, since the experimental instantiation of [10] is not described (e.g., whether dropout-adaptation training is used). Clarify the baseline configuration.
[Section V-B] The feature-importance and embedding-scaling rule is stated qualitatively ('in proportion to their reliability', 'scale linearly with the client’s reliability'); specify the exact mapping used in the experiments for reproducibility.
[Sections IV, V, VI] Typos and wording: 'alligns' (Section IV), 'its'' (Sections IV and V), 'via using' (Section VI-B), and inconsistent capitalization of 'NWDAFs'/'NWDAF server' should be corrected.
[Section VI-B] The notation a_k is used both for the Bernoulli availability indicator and in the ID definition without an explicit definition of its range; define it once in Section VI-B.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the proposed algorithm is an explicit reliability-proportional allocation heuristic, and the empirical comparison uses an external baseline with no fitted parameter renamed as a prediction. The test-set checkpoint selection in Sec. VI-B is an evaluation caveat, not a circular derivation.

full rationale

The paper contains no derived prediction that reduces to an input by construction. Algorithm 2 explicitly assigns features and embedding sizes in proportion to forecast reliability (steps 2-3), with stated assumptions that reliability remains fixed during training and testing (Sec. V-A and VI-B). The improvement claim is an empirical comparison against the external baseline [10] on a public dataset; no parameter of the proposed method is fitted to the test outcome. The protocol of storing best-performing models by test loss during training (Sec. VI-B) could inflate the reported 14.5% and 18% gains and should be corrected with a validation split, but it is a soundness concern rather than circularity because the same protocol applies to both algorithms and the algorithm itself is not defined in terms of the measured losses. There are no load-bearing self-citations or imported uniqueness claims, so the derivation chain is self-contained.

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

The algorithm introduces no new physical or mathematical entities; it relies on standard VFL components, centralized CN data management, and the assumptions listed. The main hand-chosen elements are the scaling rule, embedding budget, and reliability scenarios. All are scenario or design choices rather than fitted constants, but the central claim depends on them.

free parameters (6)

Number of selected clients K = 4
Chosen by hand for simulation; affects how features and embeddings are distributed.
Total embedding dimension = 48
Fixed communication budget hyperparameter; constrains model sizes.
Client reliability tags = {1, 2, 4, 8}
Non-linear ascending tags used to encode reliability levels and compute availability pattern IDs.
Feature importance and embedding scaling rule = linear in reliability
Design choice that maps reliability to feature importance and embedding size; the core heuristic, not derived.
Dropout distribution parameters = Beta(8,2) and Beta(10,6)
Scenario definitions for high and moderate reliability; not fitted but chosen for the experiments.
Feature importance estimator = scikit Decision Tree Regressor
Choice of importance model; no hyperparameters or validation of importance accuracy reported.

assumptions (6)

domain assumption The server is fully reliable during training and inference.
Stated in Section IV; the paper does not address server dropout, so all results depend on this simplification.
domain assumption Client reliability probabilities p_k are known or forecastable and are fixed throughout training and testing.
Section V-A asks NWDAFs to forecast reliability; Section VI-B keeps drawn probabilities fixed. Forecast error or drift is not modeled.
domain assumption Client availability in each round follows independent Bernoulli(p_k) draws.
Section VI-B specifies a_k ~ Bern(p_k); real overload events may be correlated over time.
domain assumption The 5G core can centrally store all features and reassign them arbitrarily to selected clients at low cost.
Section IV describes centralized data collection and distribution; the feature association step depends on this capability.
domain assumption Feature importance computed by a scikit Decision Tree Regressor is a faithful proxy for true predictive importance.
Section VI.B uses this estimator; if importance scores are inaccurate, the allocation to reliable clients is suboptimal.
domain assumption Zero imputation for missing embeddings is an appropriate reconstruction strategy.
Section IV states unavailable clients' embeddings are replaced with a zero vector; this choice affects both compared algorithms.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Reliable Vertical Federated Learning in 5G Core Network Architecture." pith.science (2026). https://pith.science/paper/GETJJJXW

@misc{pith2026250515244,
  author       = {Pith},
  title        = {Pith review of: Reliable Vertical Federated Learning in 5G Core Network Architecture},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GETJJJXW}},
  note         = {Machine review of arXiv:2505.15244}
}

read the original abstract

This work proposes a new algorithm to mitigate model generalization loss in Vertical Federated Learning (VFL) operating under client reliability constraints within 5G Core Networks (CNs). Recently studied and endorsed by 3GPP, VFL enables collaborative and load-balanced model training and inference across the CN. However, the performance of VFL significantly degrades when the Network Data Analytics Functions (NWDAFs) - which serve as primary clients for VFL model training and inference - experience reliability issues stemming from resource constraints and operational overhead. Unlike edge environments, CN environments adopt fundamentally different data management strategies, characterized by more centralized data orchestration capabilities. This presents opportunities to implement better distributed solutions that take full advantage of the CN data handling flexibility. Leveraging this flexibility, we propose a method that optimizes the vertical feature split among clients while centrally defining their local models based on reliability metrics. Our empirical evaluation demonstrates the effectiveness of our proposed algorithm, showing improved performance over traditional baseline methods.

Figures

Figures reproduced from arXiv: 2505.15244 by the authors.

**Figure 1.** Weighted average test loss in scenario 1 (pk ∼ Beta(8,2), averaged over 10 runs). x-axis represents the availability patterns based on assigned tags. 0 2 4 6 8 10 12 14 16 Participating NWDAFs ID/tag Sum 0.06 0.04 0.02 0.00 0.02 0.04 Difference in Weighted Loss 0.002 -0.041 -0.022 0.009 0.027 0.020 0.033 -0.034 -0.020 -0.039 -0.025 -0.019 -0.030-0.033 -0.027 Performance Comparison: Proposed Method vs. SoTA VFL Weigh… view at source ↗

**Figure 2.** Weighted test loss difference across the different ID patterns under scenario 1. C. Results and Analysis [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** Weighted average test loss in scenario 2 (pk ∼ Beta(10,6), averaged over 10 runs). x-axis represents the availability patterns based on assigned tags. 0 2 4 6 8 10 12 14 16 Participating NWDAFs ID/tag Sum 0.125 0.100 0.075 0.050 0.025 0.000 0.025 0.050 Difference in Weighted Loss 0.039 -0.003 0.027 0.002 0.043 0.020 -0.008 -0.119 -0.101 -0.063 -0.047 -0.061 -0.077 -0.029-0.030 Performance Comparison: Proposed Method… view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Weighted test loss difference across the different ID patterns under scenario 2. is relatively lower than the moderately reliable setting. Nevertheless, in this setting, our proposed algorithm still delivers approximately 14.5% lower test loss on average across all cl…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

17 extracted references · 8 canonical work pages

[10]

Robust and IP-Protecting Vertical Federated Learning against Unexpected Quitting of Parties

J. Sun et al. “Robust and IP-protecting vertical federated learning against unexpected quitting of parties,” arXiv preprint arXiv:2303.18178, 2023. [Online]. Available: https://arxiv.org/abs/2303.18178

work page Pith review arXiv 2023
[9]

Fault tolerant server- less VFL over dynamic device environment

S. Ganguli, Z. Zhou, C. G. Brinton, and D. I. Inouye, “Fault tolerant server- less VFL over dynamic device environment”, 2024. [Online]. Available: https://arxiv.org/abs/2312.16638

arXiv 2024
[1]

Feder- ated learning of deep networks using model averaging,

H. B. McMahan, E. Moore, D. Ramage, and B. Ag ¨uera y Arcas, “Feder- ated learning of deep networks using model averaging,”, 2016. [Online]. Available: http://arxiv.org/abs/1602.05629

arXiv 2016
[2]

Vertical federated learning: Concepts, advances, and chal- lenges,

Y . Liu et al. “Vertical federated learning: Concepts, advances, and chal- lenges,” *IEEE Transactions on Knowledge and Data Engineering*, vol. 36, no. 7, pp. 3615–3634, Jul. 2024. [Online]. Available: http://dx.doi.org/ 10.1109/TKDE.2024.3352628

arXiv 2024
[3]

Architecture enhancements for 5G system (5GS) to support network data analytics services,

3GPP TS 23.288, “Architecture enhancements for 5G system (5GS) to support network data analytics services,” V18.6.0, Section 5.3: ”Federated Learning (FL) among multiple NWDAFs”, June 2024

work page 2024
[4]

3GPP TR 23.700-84, ”Study on core network enhanced support for artificial intelligence (AI)/machine learning (ML)”, V2.0.0, Sep. 2024. [3rd Generation Partnership Project (3GPP), Technical Specification Group Services and System Aspects, Release 19]

work page 2024
[5]

Federated multi-task learning for competing constraints

T. Li, S. Hu, A. Beirami, and V . Smith, “Federated multi-task learning for competing constraints”, 2020. [Online]. Available: https://arxiv.org/abs/ 2012.04221

arXiv 2020
[6]

On the convergence of federated optimization in heterogeneous networks

A. K. Sahu et al. “On the convergence of federated optimization in heterogeneous networks”, 2018. [Online]. Available: http://arxiv.org/abs/ 1812.06127

arXiv 2018

Show all 17 references

[7]

M. M. Amiri and D. G ¨und¨uz, ”Federated Learning Over Wireless Fading Channels,” in IEEE Transactions on Wireless Communications, vol. 19, no. 5, pp. 3546-3557, May 2020, doi: 10.1109/TWC.2020.2974748

2020
[8]

Zhang et al., ”TreeCSS: An Efficient Framework for Vertical Federated Learning”, 2024, Available: https://arxiv.org/abs/2408.01691,

Q. Zhang et al., ”TreeCSS: An Efficient Framework for Vertical Federated Learning”, 2024, Available: https://arxiv.org/abs/2408.01691,

2024 arXiv
[11]

A distributed NWDAF architecture for federated learning in 5G,

Y . Jeon et al. “A distributed NWDAF architecture for federated learning in 5G,” *2022 IEEE International Conference on Consumer Electronics (ICCE)*, Las Vegas, NV , USA, 2022, pp. 1-2, doi: 10.1109/ICCE53296.2022.9730220

2022
[12]

Architecture enhancements for 5G system (5GS) to support network data analytics services,

3GPP TS 23.288, “Architecture enhancements for 5G system (5GS) to support network data analytics services,” V17.8.0, March 2023

2023
[13]

Architecture enhancements for 5G system (5GS) to support network data analytics services,

3GPP TS 23.288, “Architecture enhancements for 5G system (5GS) to support network data analytics services,” V18.6.0, Section 5A: ”Data Collection Coordination and Delivery Functional Description”, June 2024

2024
[14]

Architecture enhancements for 5G system (5GS) to sup- port network data analytics services,

3GPP TS 23.288, “Architecture enhancements for 5G system (5GS) to sup- port network data analytics services,” V18.6.0, Section 4.2.1: ”Analytics Data Repository Function Functional Description”, June 2024

2024
[15]

A unified approach to interpreting model pre- dictions,

S. Lundberg and S.-I. Lee, “A unified approach to interpreting model pre- dictions,” *arXiv preprint arXiv:1705.07874*, 2017. [Online]. Available: https://arxiv.org/abs/1705.07874

2017 arXiv
[16]

Flower: A friendly federated learning research frame- work

D. J. Beutel et al. “Flower: A friendly federated learning research frame- work”, 2020. [Online]. Available: https://arxiv.org/abs/2007.14390

2020 arXiv
[17]

Pedregosa et al., ”scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, vol

F. Pedregosa et al., ”scikit-learn: Machine Learning in Python,” Journal of Machine Learning Research, vol. 12, pp. 2825-2830, 2011

2011

Pith tools

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

[1] [10]

Robust and IP-Protecting Vertical Federated Learning against Unexpected Quitting of Parties

J. Sun et al. “Robust and IP-protecting vertical federated learning against unexpected quitting of parties,” arXiv preprint arXiv:2303.18178, 2023. [Online]. Available: https://arxiv.org/abs/2303.18178

work page Pith review arXiv 2023

[2] [9]

Fault tolerant server- less VFL over dynamic device environment

S. Ganguli, Z. Zhou, C. G. Brinton, and D. I. Inouye, “Fault tolerant server- less VFL over dynamic device environment”, 2024. [Online]. Available: https://arxiv.org/abs/2312.16638

arXiv 2024

[3] [1]

Feder- ated learning of deep networks using model averaging,

H. B. McMahan, E. Moore, D. Ramage, and B. Ag ¨uera y Arcas, “Feder- ated learning of deep networks using model averaging,”, 2016. [Online]. Available: http://arxiv.org/abs/1602.05629

arXiv 2016

[4] [2]

Vertical federated learning: Concepts, advances, and chal- lenges,

Y . Liu et al. “Vertical federated learning: Concepts, advances, and chal- lenges,” *IEEE Transactions on Knowledge and Data Engineering*, vol. 36, no. 7, pp. 3615–3634, Jul. 2024. [Online]. Available: http://dx.doi.org/ 10.1109/TKDE.2024.3352628

arXiv 2024

[5] [3]

Architecture enhancements for 5G system (5GS) to support network data analytics services,

3GPP TS 23.288, “Architecture enhancements for 5G system (5GS) to support network data analytics services,” V18.6.0, Section 5.3: ”Federated Learning (FL) among multiple NWDAFs”, June 2024

work page 2024

[6] [4]

3GPP TR 23.700-84, ”Study on core network enhanced support for artificial intelligence (AI)/machine learning (ML)”, V2.0.0, Sep. 2024. [3rd Generation Partnership Project (3GPP), Technical Specification Group Services and System Aspects, Release 19]

work page 2024

[7] [5]

Federated multi-task learning for competing constraints

T. Li, S. Hu, A. Beirami, and V . Smith, “Federated multi-task learning for competing constraints”, 2020. [Online]. Available: https://arxiv.org/abs/ 2012.04221

arXiv 2020

[8] [6]

On the convergence of federated optimization in heterogeneous networks

A. K. Sahu et al. “On the convergence of federated optimization in heterogeneous networks”, 2018. [Online]. Available: http://arxiv.org/abs/ 1812.06127

arXiv 2018