Pith. sign in

REVIEW 3 major objections 4 minor 1 cited by

Optimal Transceiver Design in Over-the-Air Federated Distillation

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read Over-the-air federated distillation replaces model uploads with aggregated soft predictions, and its beamforming design is proved to have zero relaxation gap.

desk verdict A useful over-the-air FD framework with a clean power-control result, but the zero-optimality-gap theorem in Proposition 2 rests on a false inequality. read the letter →

arxiv 2507.15256 v1 pith:M5WSKBKD submitted 2025-07-21 eess.SP cs.AI

Zihao Hu (1) , Jia Yan (2) , Ying-Jun Angela Zhang (1) , Jun Zhang (3) , Khaled B. Letaief (3) ((1) The Chinese University of Hong Kong , (2) The Hong Kong University of Science , Technology (Guangzhou) , (3) The Hong Kong University of Science
show 1 more author
Technology)
This is my paper · ORCID
classification eess.SPcs.AI
keywords federateddistillationover-the-aircomputationtransceiverdesignbeamformingsemidefiniterelaxationconvergencerateknowledgewirelessedgelearning
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 a wireless federated distillation scheme in which devices upload only per-class averaged soft predictions, called their knowledge, and the parameter server aggregates these predictions over the air by exploiting channel superposition, avoiding transmission of large model parameters. It derives an upper bound on the expected gradient norm that serves as a convergence-rate proxy, then formulates a transceiver optimization problem over transmit powers, aggregation post-processing scalars, and receiver beamforming. The authors claim that, given the receiving strategy, the transmit powers and scalars have closed-form optimal solutions, and that the semidefinite relaxation of the beamforming problem is tight because its optimal solution is rank-one. If these claims hold, over-the-air federated distillation can approach error-free distillation accuracy with a fraction of the communication load of conventional federated learning on large models.

What carries the argument

The engine of the argument is the convergence-rate upper bound in Theorem 1, in which the expected squared gradient norm is bounded by terms involving $\Phi_{1,i,t}$, the signal misalignment caused by channel fading and power variation across devices, and $\Phi_{2,i,t}$, the noise passed through the post-processing scalar. Minimizing that bound decouples into a closed-form power-and-estimator step and a beamforming problem that is reformulated, via slack variables $e_k$ and matrices $W_t=w_t w_t^H$, as the semidefinite program (P6). The rank-one proof then constructs a feasible rank-one point from the eigen-decomposition $W_t^* = V\Lambda V^H$ by choosing $\hat v = \sum_{r=1}^R \sqrt{\Lambda_r} v_r$ and using inequality (53) to argue that this rank-one point has no larger objective.

What would settle it

Run the relaxed SDP (P6) on a channel realization where the device with the smallest value of $|w^H h_{j,t}|/\hat{q}_{j,t}^k$ differs across classes; if the optimal value of (P6) is strictly lower than the best achievable value under the rank-one constraint, then the zero-gap claim is false.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that transceiver design for over-the-air federated distillation is not merely a good heuristic but is globally optimal. For a fixed receiver beamforming vector $w_t$, the optimal transmit equalization factors $P_{i,t}^{k*}$ and post-processing scalars $\lambda_t^{k*}, a_{i,t}^{k*}$ are given in closed form in Proposition 1. For the remaining beamforming problem, the paper formulates a semidefinite relaxation and proves in Proposition 2 that the relaxed optimum $W_t^*$ is always rank-one, so the recovered beamforming vector $w_t^*$ solves the original non-convex problem with zero optimality gap. Together with the convergence-rate expression of Theorem 1, this yields a per-round optimization algorithm whose objective is the learning performance itself rather than a proxy such as mean squared error.

Load-bearing premise

The proof of Proposition 2 relies on an inequality that does not hold for arbitrary complex numbers, and it assumes that the same device is the bottleneck for every class, so the rank-one construction may not actually reach the relaxed optimum.

Editorial extensions

If this is right

  • Each of the three design variables, transmit power, aggregation estimator, and receiver beamforming, has a per-round optimal form, and the beamforming step reduces to a convex SDP whose solution is rank-one.
  • The optimal design depends only on channel state, local class counts, and knowledge statistics, not on learning constants such as $L_1$ and $L_2$, so the same algorithm can be run without estimating those constants.
  • Communication per round scales with $K^2$, the number of classes squared, rather than with model size, which is the paper's route to applying federated distillation to very large models.
  • In the paper's simulations over FMNIST and CIFAR-10 with a 5-antenna server and 50 devices, the design approaches error-free FD accuracy and outperforms uniform beamforming and MSE-minimizing baselines in both IID and non-IID splits.
  • The per-round complexity $O(N^3+M)$ lets the server recompute the design at the start of each fading block, avoiding per-device channel state information at the transmitter.

Reading between the lines

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

  • A reader should not treat the zero-gap proof as settled: the inequality in (53), $|\sum_r z_r\sqrt{\Lambda_r}|^2 \ge \sum_r \Lambda_r |z_r|^2$, is not valid for arbitrary complex $z_r$, so the constructed rank-one matrix may be strictly worse than the relaxed optimum.
  • A direct test is to generate channels where class 1's bottleneck device differs from class 2's bottleneck device; if the relaxed SDP beats every rank-one feasible point in such a setting, the optimality gap is real and a rank-restricted or regularized beamforming step is needed.
  • If the zero-gap claim survives, the same slack-variable SDR structure may apply to other multi-class over-the-air aggregation problems, such as multi-task or multi-modal federated distillation.
  • The convergence-rate objective being free of $L_1$ and $L_2$ suggests the design could be implemented without a learning-performance model, but it also leaves open how much accuracy is lost by optimizing a bound rather than the true convergence rate.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 4 minor

Summary. The paper proposes an over-the-air federated distillation framework in which wireless devices upload per-class averaged soft predictions, aggregated at a multi-antenna parameter server by over-the-air computation. It derives a convergence bound for the FD training process, formulates a transceiver design problem, obtains closed-form optimal transmit equalization factors and post-processing scalars, and proposes an SDR-based receiver beamforming method. The authors claim that the SDR relaxation has zero optimality gap (Proposition 2) and support this with numerical experiments.

Significance. If the results hold, the paper makes a useful contribution: it is the first transceiver design for over-the-air federated distillation, with a nontrivial convergence analysis and a low-complexity alternating algorithm. The closed-form power control and the SDR formulation are natural, and the numerical comparisons show meaningful communication-overhead reductions. However, the central theoretical guarantee—the zero optimality gap for the beamforming relaxation—rests on a proof that is not valid as written, so the paper's main optimality claim is not established by the current manuscript.

major comments (3)
  1. [Appendix C, Eq. (53)] The inequality |∑_r z_r √Λ_r|² ≥ ∑_r Λ_r |z_r|² is stated to follow from Cauchy–Schwarz, but Cauchy–Schwarz gives the opposite bound. For z1=1, z2=−1 and Λ1=Λ2=1/2, the left side is 0 and the right side is 1, so the inequality fails. Since (53) is used to prove that the constructed matrix W_hat satisfies the constraints (54) and that Ω(W_hat) ≤ Ω(W*), the proof of Proposition 2 does not establish feasibility or optimality of the rank-one candidate.
  2. [Appendix C, Eq. (49)] The 'without loss of generality' assumption that device 1 attains equality in every per-class constraint is not justified. Lemma C.2 only guarantees at least one equality per constraint set, and the equality-achieving device index can vary with k; the subsequent construction using a single h1,t and z1,r for all k is therefore unsupported.
  3. [Section V-B and Fig. 2] The numerical observation that the second eigenvalue of W*_t is near zero is only evidence for the tested random instances and is computed at finite precision. It does not replace the proof of Proposition 2, and the abstract's 'no optimality gap' claim therefore remains unproven.
minor comments (4)
  1. [Appendix C, Eq. (52)] The standard basis vectors used to complete the orthonormal basis should be indexed up to N, not K, since the ambient dimension is N; as written, the construction is undefined when K ≠ N.
  2. [Lemma C.1] The statement that the objective is coercive because it approaches infinity when e_k approaches infinity is imprecise; the objective decreases without bound as e_k → −∞, though feasibility restricts e_k from below. A correct compactness argument is available and should be stated.
  3. [Proposition 1, Eq. (20)] The notation (w^H h)^H in the numerator is opaque; writing h^H w explicitly would make the formula easier to read.
  4. [Footnote 4] The remark that the optimal transceiver design is independent of L1 and L2 is correct for the power-control and beamforming steps, but the convergence bound in Theorem 1 still depends on these constants; the wording could be sharpened to avoid overstatement.

Circularity Check

0 steps flagged · score 1.0 of 10

No circularity: the optimal transceiver design is derived from an independently derived convergence upper bound; the only self-citation [1] is a non-load-bearing conference-preprint note.

full rationale

This paper's derivation chain is not circular. Theorem 1 derives an upper bound on the expected gradient norm from Assumptions 1-3 (Lipschitz continuity of loss gradients, Lipschitz continuity of the model mapping, and bounded gradient norms); no target accuracy or fitted simulation result is substituted into the bound. Proposition 1 minimizes the resulting bound in closed form by enforcing the alignment conditions in Eq. (41), and the inequality lower bound in Eq. (40) is an algebraic reduction rather than an equivalence imposed by definition. Proposition 2 is an SDR-tightness proof: the authors construct a rank-one candidate and compare objective values. The Skeptic's objection concerns Eq. (53), where the claimed Cauchy-Schwarz inequality has the wrong direction for complex coefficients; that is a proof gap or correctness risk, not circularity, because the conclusion is not assumed in the premise and is not produced by a fitted input. The only self-citation is [1], a VTC conference version of this work, cited only as 'presented in part'; it is not load-bearing for any theorem. Numerical evaluation is against external benchmarks (error-free FD/FedAvg, uniform beamforming, MSE minimization, pruning/QAT baselines), so the empirical claims are self-contained. Accordingly, no circular step is identified; the score of 1 reflects only the presence of a non-load-bearing self-citation.

Assumptions & free parameters 0 free parameters · 7 assumptions · 0 invented entities

The convergence bound rests on standard Lipschitz and boundedness assumptions plus the block-fading and error-free downlink assumptions. The proof of Proposition 2 additionally asserts an inequality that is not generally true; this is the main unsupported axiom. No parameters are fitted to data.

assumptions (7)
  • domain assumption Assumption 1: Gradients of the local loss functions are L1-Lipschitz continuous (Eq. 14).
    Standard smoothness condition used throughout the convergence proof in Appendix A.
  • domain assumption Assumption 2: Model function G_theta is L2-Lipschitz continuous in theta (Eq. 15).
    Used to bound the distillation regularizer term in Appendix A.
  • domain assumption Assumption 3: Gradient norm is uniformly bounded by S (Eq. 16).
    Used to bound the loss decrease term B1 in Appendix A.
  • domain assumption The empirical local loss function F_i(theta) is upper bounded by f_i,max.
    Required for the telescoping sum in the proof of Theorem 1; stated before Theorem 1.
  • domain assumption Block fading channel, constant within one training round, and perfect CSI at the PS.
    System model in Section II.C; the per-round optimization relies on knowing the channels.
  • domain assumption Downlink broadcast from the PS to WDs is error-free.
    Stated in Section II.A; all WDs receive the same estimated knowledge without error.
  • ad hoc to paper Inequality |sum_r z_r sqrt(Lambda_r)|^2 >= sum_r Lambda_r |z_r|^2 used to construct a rank-one solution.
    Asserted in Appendix C, Eq. (53). It is not generally true for complex z_r, so the proof of Proposition 2 is invalid.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Optimal Transceiver Design in Over-the-Air Federated Distillation." pith.science (2026). https://pith.science/paper/M5WSKBKD

@misc{pith2026250715256,
  author       = {Pith},
  title        = {Pith review of: Optimal Transceiver Design in Over-the-Air Federated Distillation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/M5WSKBKD}},
  note         = {Machine review of arXiv:2507.15256}
}
read the original abstract

The rapid proliferation and growth of artificial intelligence (AI) has led to the development of federated learning (FL). FL allows wireless devices (WDs) to cooperatively learn by sharing only local model parameters, without needing to share the entire dataset. However, the emergence of large AI models has made existing FL approaches inefficient, due to the significant communication overhead required. In this paper, we propose a novel over-the-air federated distillation (FD) framework by synergizing the strength of FL and knowledge distillation to avoid the heavy local model transmission. Instead of sharing the model parameters, only the WDs' model outputs, referred to as knowledge, are shared and aggregated over-the-air by exploiting the superposition property of the multiple-access channel. We shall study the transceiver design in over-the-air FD, aiming to maximize the learning convergence rate while meeting the power constraints of the transceivers. The main challenge lies in the intractability of the learning performance analysis, as well as the non-convex nature and the optimization spanning the whole FD training period. To tackle this problem, we first derive an analytical expression of the convergence rate in over-the-air FD. Then, the closed-form optimal solutions of the WDs' transmit power and the estimator for over-the-air aggregation are obtained given the receiver combining strategy. Accordingly, we put forth an efficient approach to find the optimal receiver beamforming vector via semidefinite relaxation. We further prove that there is no optimality gap between the original and relaxed problem for the receiver beamforming design. Numerical results will show that the proposed over-the-air FD approach achieves a significant reduction in communication overhead, with only a minor compromise in testing accuracy compared to conventional FL benchmarks.

Figures

Figures reproduced from arXiv: 2507.15256 by the authors.

Figure 1
Figure 1. The proposed over-the-air FD framework. of the WDs, estimator for over-the-air aggregation, and the receiver combining scheme, by maximizing the con￾vergence rate of the considered FD system. • To tackle the challenging non-convex transceiver design problem, we derive the closed-form solutions of the optimal transmit equalization factors and post-processing scalars, given the receiver beamforming strategy. • We desi… view at source ↗
Figure 4
Figure 4. Average testing accuracy versus the distillation we [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Analytical convergence gap and average global testing accuracy versus the number of antennas at the PS. C. Effect of Distillation Weight We then study the impact of the distillation weight γ on the learning performance. Notice that γ in the local loss function of each WD balances the general training loss (f(·) in (1)), and the distillation loss. In [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figures from the paper (2 more)
Figure 6
Figure 6. Figure 6: Average global testing accuracy of ResNet-18 over CIFAR-10 in the IID setting. • (Over-the-air FL with) 90%-Sparse Rate [9]: Each WD performs local model training for E = 5 iterations by stochastic gradient descent. Then, each WD sparsifies 90% of its local gradient an…
Figure 7
Figure 7. Figure 7: Average global testing accuracy of ResNet-18 over CIFAR-10 in the non-IID setting. the highest testing accuracy among the other benchmarks under the restricted time consumption (i.e., smaller than 60000 seconds). Then, we see that the uniform beamforming ap￾proach achi…

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Communication-Learning Co-Design for Differentially Private Over-the-Air Federated Distillation

    cs.IT 2025-08 reject novelty 5.0 of 10

    Closed-form transmit and training-round designs are derived for differentially private over-the-air federated distillation, claiming a better learning-privacy-communication trade-off than federated learning.

Reference graph

Works this paper leans on

32 extracted references · 29 canonical work pages · cited by 1 Pith paper

  1. [1]

    Communication-Learning Co-Design for Over-the-Air Fede rated Dis- tillation,

    Z. Hu, J. Y an, Y .-J. A. Zhang, J. Zhang, and K. B. Letaief, “Communication-Learning Co-Design for Over-the-Air Fede rated Dis- tillation,” 2024 IEEE 99th V ehicular Technology Conference (VTC2024- Spring), Singapore, Jun. 2024

    work page 2024

  2. [2]

    A Fai r and Efficient Federated Learning Algorithm for Autonomous Driv ing,

    X. Tang, J. Zhang, Y . Fu, C. Li, N. Cheng, and X. Y uan, “A Fai r and Efficient Federated Learning Algorithm for Autonomous Driv ing,” 2023 IEEE 98th V ehicular Technology Conference (VTC2023-Fall) , Hong Kong, Oct. 2023, pp. 1-5

    work page 2023

  3. [3]

    Designing applications for collaborative learning in vir tual reality,

    M. Tataru, S. Berzescu, S. V ert, V . Mihaescu, R. Stamatoiu, and R. V asiu, “Designing applications for collaborative learning in vir tual reality,” in Proc. Int. Symp. Electron. Telecommun. (ISETC), Nov. 2022, pp. 1–4

    work page 2022

  4. [4]

    The roadmap to 6G: AI empowered wireless networks,

    K. B. Letaief, W. Chen, Y . Shi, J. Zhang, and Y .-J. Zhang, “ The roadmap to 6G: AI empowered wireless networks,” IEEE Communications Mag- azine, vol. 57, no. 8, pp. 84-90, Aug. 2019

    work page 2019

  5. [5]

    Com munication- efficient edge AI: Algorithms and systems,

    Y . Shi, K. Y ang, T. Jiang, J. Zhang, and K. B. Letaief, “Com munication- efficient edge AI: Algorithms and systems,” IEEE Commun. Surveys Tuts., vol. 22, no. 4, pp. 2167–2191, 4th Quart. 2020

    work page 2020

  6. [6]

    Communication-efficient learning of deep networks from de centralized data,

    B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. Y . Ar cas, “Communication-efficient learning of deep networks from de centralized data,” in Proc. 20th Int. Conf. Artif. Intell. Stat., vol. 54, Apr. 2017, pp.1273–1282

    work page 2017

  7. [7]

    Broadband Analog Aggregat ion for Low-Latency Federated Edge Learning,

    G. Zhu, Y . Wang, and K. Huang, “Broadband Analog Aggregat ion for Low-Latency Federated Edge Learning,” in IEEE Transactions on Wireless Communications, vol. 19, no. 1, pp. 491-506, Jan. 2020

    work page 2020

  8. [8]

    Over- the-air federated learning and optimization,

    J. Zhu, Y . Shi, Y . Zhou, C. Jiang, W. Chen, and K. B. Letaief , “Over- the-air federated learning and optimization,” IEEE Internet of Things Journal, vol. 11, no. 11, pp. 16996-17020, May 2024

    work page 2024

Show all 32 references
  1. [9]

    Federated Learning Over W ireless Fading Channels,

    M. M. Amiri, and, D. G¨ und¨ uz, “Federated Learning Over W ireless Fading Channels,” in IEEE Transactions on Wireless Communications, vol. 19, no. 5, pp. 3546-3557, May 2020

    2020

  2. [10]

    Joint Receiver Desi gn and User Scheduling for Over-the-Air Aggregation in Federated Lear ning,

    F. Zhang, J. Wan, K. Wang, and Q. Wu, “Joint Receiver Desi gn and User Scheduling for Over-the-Air Aggregation in Federated Lear ning,” 2023 5th International Conference on Communications, Informat ion System and Computer Engineering (CISCE), Guangzhou, China, Apr. 2023, pp. 20-25

    2023

  3. [11]

    Federated Learnin g via Over-the- Air Computation,

    K. Y ang, T. Jiang, Y . Shi and Z. Ding, “Federated Learnin g via Over-the- Air Computation,” in IEEE Transactions on Wireless Communications, vol. 19, no. 3, pp. 2022-2035, Mar. 2020

    2022

  4. [12]

    Over-the-Air Compu tation Empowered Federated Learning: A Joint Uplink-Downlink Des ign,

    D. Zhang, M. Xiao, and M. Skoglund, “Over-the-Air Compu tation Empowered Federated Learning: A Joint Uplink-Downlink Des ign,” 2023 IEEE 98th V ehicular Technology Conference (VTC2023-F all), Hong Kong, Oct. 2023, pp. 1-7

    2023

  5. [13]

    Joint Beamforming and Learning Rat e Opti- mization for Over-the-Air Federated Learning,

    M. Kim, and D. Park, “Joint Beamforming and Learning Rat e Opti- mization for Over-the-Air Federated Learning,” in IEEE Transactions on V ehicular Technology, vol. 72, no. 10, pp. 13706-13711, Oct. 2023

    2023

  6. [14]

    Harnessing the power of llms in practice: A survey on chatgpt and beyond,

    J. Y ang, H. Jin, R. Tang, X. Han, Q. Feng, H. Jiang, S. Zhon g, B. Yin, and X. Hu, “Harnessing the power of llms in practice: A survey on chatgpt and beyond,” ACM Transactions on Knowledge Discovery from Data, vol. 18, no. 6, pp. 1-32, Jul. 2024

    2024

  7. [15]

    Deep Compr ession for Efficient and Accelerated Over-the-Air Federated Learn ing,

    F. M. A. Khan, H. Abou-Zeid, and S. A. Hassan, “Deep Compr ession for Efficient and Accelerated Over-the-Air Federated Learn ing,” in IEEE Internet of Things Journal, vol. 11, no. 15, pp. 25802-25817, 1 Aug.1, 2024

    2024

  8. [16]

    Communication-efficient on-device machine learning: fed erated distil- lation and augmentation under non-IID private data,

    E. Jeong, S. Oh, H. Kim, J. Park, M. Bennis, and S. Kim, “Communication-efficient on-device machine learning: fed erated distil- lation and augmentation under non-IID private data,” in Proc. NIPS, 2018

    2018

  9. [17]

    Large scale distributed neural network training t hrough online distillation,

    R. Anil, G. Pereyra, A. Passos, R. Ormandi, G. E. Dahl, an d G. E. Hinton, “Large scale distributed neural network training t hrough online distillation,” arXiv preprint arXiv:1804.03235, 2018

    2018 arXiv

  10. [18]

    CFD: Communication-Efficient Federated Distillation via Soft- Label Quanti- zation and Delta Coding,

    F. Sattler, A. Marban, R. Rischke, and W. Samek, “CFD: Communication-Efficient Federated Distillation via Soft- Label Quanti- zation and Delta Coding,” in IEEE Transactions on Network Science and Engineering, vol. 9, no. 4, pp. 2025-2038, 1 Jul.-Aug. 2022

    2025

  11. [19]

    Selective knowledge shari ng for privacy- preserving federated distillation without a good teacher,

    J. Shao, F. Wu, and J. Zhang, “Selective knowledge shari ng for privacy- preserving federated distillation without a good teacher, ” arXiv preprint arXiv:2304.01731, 2023

    2023 arXiv

  12. [20]

    Communication Chann el-Aware Federated Distillation For Massive MIMO Systems,

    Y . Mu, N. Garg, and T. Ratnarajah, “Communication Chann el-Aware Federated Distillation For Massive MIMO Systems,” 2023 IEEE 24th International W orkshop on Signal Processing Advances in Wi reless Communications (SPAWC), Shanghai, China, Sept. 2023, pp. 436-460

    2023

  13. [21]

    Cooperative Learni ng VIA Fed- erated Distillation OVER Fading Channels,

    J.-H. Ahn, O. Simeone, and J. Kang, “Cooperative Learni ng VIA Fed- erated Distillation OVER Fading Channels,” 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICA SSP), Barcelona, Spain, May 2020, pp. 8856-8860

    2020

  14. [22]

    Joint Chann el Esti- mation and Feedback for mm-Wave System Using Federated Lear ning,

    L. Zhao, H. Xu, Z. Wang, X. Chen, and A. Zhou, “Joint Chann el Esti- mation and Feedback for mm-Wave System Using Federated Lear ning,” in IEEE Communications Letters, vol. 26, no. 8, pp. 1819-1823, Aug. 2022

    2022

  15. [23]

    Federated Generative Ad versarial Networks based Channel Estimation,

    Y . Guo, Z. Qin, and O. A. Dobre, “Federated Generative Ad versarial Networks based Channel Estimation,” 2022 IEEE International Confer- ence on Communications W orkshops (ICC W orkshops), Seoul, Republic of Korea, May 2022, pp. 61-66

    2022

  16. [24]

    The Best of Both Worlds: Accurat e Global and Personalized Models through Federated Learning with Data- Free Hyper- Knowledge Distillation,

    H. Chen, and H. Vikalo, “The Best of Both Worlds: Accurat e Global and Personalized Models through Federated Learning with Data- Free Hyper- Knowledge Distillation,” arXiv preprint arXiv:2301.08968, 2023

    2023 arXiv

  17. [25]

    On the conver gence of step decay step-size for stochastic optimization,

    X. Wang, S. Magn´ usson, and M. Johansson, “On the conver gence of step decay step-size for stochastic optimization,” Advances in Neural Information Processing Systems, vol. 34, pp. 14226-14238, 2021

    2021

  18. [26]

    Revisit ing Analog Over-the-Air Machine Learning: The Blessing and Cur se of Interference,

    H. H. Y ang, Z. Chen, T. Q. S. Quek, and H. V . Poor, “Revisit ing Analog Over-the-Air Machine Learning: The Blessing and Cur se of Interference,” IEEE J. Sel. Topics Signal Process., vol. 16, no. 3, pp. 406–419, Apr. 2022

    2022

  19. [27]

    Sem idef- inite Relaxation of Quadratic Optimization Problems,

    Z.-Q. Luo, W.-K. Ma, A. M.-C. So, Y . Y e, and S. Zhang, “Sem idef- inite Relaxation of Quadratic Optimization Problems,” in IEEE Signal Processing Magazine, vol. 27, no. 3, pp. 20-34, May 2010

    2010

  20. [28]

    Grant, and S

    M. Grant, and S. Boyd, CVX: Matlab software for discipli ned convex programming, version 2.0 beta. http://cvxr.com/cvx, Sept . 2013

    2013

  21. [29]

    Reconfigurable Inte lligent Sur- face Enabled Federated Learning: A Unified Communication-L earning Design Approach,

    H. Liu, X. Y uan, and Y .-J. A. Zhang, “Reconfigurable Inte lligent Sur- face Enabled Federated Learning: A Unified Communication-L earning Design Approach,” in IEEE Transactions on Wireless Communications, vol. 20, no. 11, pp. 7595-7609, Nov. 2021

    2021

  22. [30]

    Wireless Communications, Principles a nd Practice,

    T. Rappaport, “Wireless Communications, Principles a nd Practice,” 2nd ed., Printice-Hall Inc., 2002

    2002

  23. [31]

    Federated learning on non-iid data silos: An experimental study,

    Q. Li, Y . Diao, Q. Chen, and B. He, “Federated learning on non-iid data silos: An experimental study,” In 2022 IEEE 38th international conference on data engineering (ICDE), pp. 965-978, May 2022

    2022

  24. [32]

    IEEE 802.11ac: 5th g eneration wifi networking,

    A. Haque, P . Kumar, and A. K. Singh, “IEEE 802.11ac: 5th g eneration wifi networking,” W orld Applied Programming, vol. 2, no. 4, pp. 235–241, 2012

    2012

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.