Pith. sign in

REVIEW 3 major objections 5 minor 1 cited by

SplitFedZip: Learned Compression for Data Transfer Reduction in Split-Federated Learning

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read Learned rate-distortion codecs at the split points of a split-federated network reduce transferred data by at least three orders of magnitude while preserving segmentation accuracy.

desk verdict Sound SplitFed compression study with a strong HAM10K result, but the conclusion overclaims on Blastocyst and the comparison table has accounting problems. read the letter →

arxiv 2412.17150 v1 pith:NVEW24U4 submitted 2024-12-18 cs.LG

Chamani Shiranthika , Hadi Hadizadeh , Parvaneh Saeedi , Ivan V. Bajić This is my paper · ORCID
classification cs.LG
keywords splitfederatedlearninglearnedcompressionrate-distortionoptimizationmedicalimagesegmentationcommunicationefficiencygradientfeatureBjøntegaardDelta
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

SplitFedZip is a way to compress the tensors that travel between client and server in split-federated learning. The paper places learned neural compressors at the two split points of a split U-Net and trains them together with the segmentation model, so that both forward feature maps and backward gradients are compressed before transmission and reconstructed on arrival. On two medical image segmentation datasets, Blastocyst and HAM10K, the paper reports that this reduces transferred data by at least three orders of magnitude relative to no compression while keeping the mean Jaccard index at the uncompressed baseline (0.892 on both datasets). A single weighting parameter $\lambda$ controls the trade-off between bit-rate and accuracy, and a suitable value reaches baseline accuracy at compression ratios of roughly $10^{3.6}$ to $10^{4.4}$. The authors describe this as the first end-to-end rate-distortion optimized compression scheme for SplitFed learning.

What carries the argument

The machinery is a pair of learned codecs attached at the split points of the network, with separate codecs for forward features and backward gradients. Each codec is trained jointly with the segmentation model using the loss $\mathcal{L} = \mathcal{L}_r + \lambda(\mathcal{L}_{\text{Dice}} + \mathcal{L}_{\text{mse}})$, where $\mathcal{L}_r$ is the rate estimate from the codec's entropy model and $\lambda$ trades bit-rate against a distortion term that includes both the Dice segmentation loss and reconstruction mean squared error. Federated averaging is applied to codec parameters as well as model parameters, so the compressors adapt to the changing statistics of the tensors they compress over global epochs. Two codec architectures are tested: a hyperprior autoencoder with generalized divisive normalization and an attention-based model with discretized Gaussian mixture likelihoods, both borrowed from learned image compression.

What would settle it

Recompute the data transfer for both methods from a single explicit formula that includes the number of training samples, local epochs, global epochs, and the number of transmissions per sample, using identical settings for SplitFedZip and the baseline; if the reduction ratio does not remain in (or near) the 3.2x-5000x range, the central communication-savings claim is refuted.

Watch

Extended reading notes

Core claim

The central claim is that learned rate-distortion codecs can be inserted into the SplitFed training loop without disturbing convergence: the codecs are updated by the same federated averaging that updates the network, and they learn to match the statistics of features and gradients as training proceeds. The paper demonstrates that both a feature-only scheme and a feature-plus-gradient scheme let the final model match the no-compression segmentation accuracy, with the feature-plus-gradient scheme compressing traffic in both directions. It also shows that an attention-based codec with Gaussian mixture entropy modeling saves roughly 69% bit-rate over a hyperprior autoencoder in the feature-only scheme and 24% in the feature-plus-gradient scheme, measured by Bjøntegaard $\Delta$ on weighted average, with only about 0.002-0.009 change in mean Jaccard index. The discovery, as the authors state it, is that SplitFed communication can be cut by orders of magnitude without sacrificing task performance by making compression a learned part of the federated model.

Load-bearing premise

The load-bearing premise is that the data-transfer figures for SplitFedZip and for the comparison method count the same number of training samples, local/global epochs, and forward/backward transmissions, so that the reported 3.2x-5000x reduction is an apples-to-apples comparison.

Editorial extensions

If this is right

  • SplitFed training becomes feasible on very low-bandwidth links by choosing a $\lambda$ that gives a 1000x or larger reduction in bytes transferred, at a modest, tunable cost in accuracy.
  • The compression codecs are part of the federated aggregation, so communication savings require no changes to the SplitFed privacy model or to the split-network training loop.
  • Compressing gradients as well as features addresses the backward pass, which is often the dominant communication cost in split learning.
  • The comparison metrics (BD-BPP, BD-MJI) give a concrete way to rank future codec designs for SplitFed, with the attention-based codec setting a new efficiency bar on the two test datasets.

Reading between the lines

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

  • The same split-point codec insertion should transfer to other architectures that shuttle intermediate tensors, such as split transformers or pipeline-parallel models, where the tensors are even larger and the potential savings correspondingly greater; this is an extrapolation the paper does not test.
  • The occasional accuracy improvements seen on HAM10K at moderate compression suggest a possible regularizing effect, but the paper does not isolate it; a controlled ablation with varying $\lambda$ schedules could clarify when compression helps rather than merely preserves accuracy.
  • The comparison with the prior SplitFed compression method rests on data-transfer formulas that omit explicit epoch multipliers; an independent re-computation using identical training-configuration assumptions would test whether the claimed 3.2x-5000x advantage holds under a fully consistent accounting.
  • A practical deployment would need to weigh the added compute and memory of the codecs themselves; the paper reports bit-rate and accuracy but not wall-clock time or energy, so the end-to-end efficiency gain remains to be quantified.
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 / 5 minor

Summary. The paper proposes SplitFedZip, a method that inserts learned compression codecs (a Ball\'e-style hyperprior AE and the Cheng2020 attention model) at the two split points of a split U-Net in Split-Federated learning, compressing both forward features and backward gradients. Training minimizes a rate term plus a λ-weighted combination of Dice and MSE distortion. Experiments on Blastocyst and HAM10K segmentation report rate-accuracy curves, BD-MJI/BD-BPP comparisons between codecs, data-transfer (DT) reductions versus no compression, and a comparison with the earlier SplitFed compression method of Ayad et al. The paper claims transferred-data reduction of at least three orders of magnitude on both datasets without hindering final model performance.

Significance. The idea of applying end-to-end rate-distortion optimized learned compression specifically to the features and gradients flowing through SplitFed split points is a meaningful and timely contribution, and the paper is honest in presenting R-A curves as empirical sweeps over λ with bit-rates measured from actual bitstreams. The open-source implementation, use of standard CompressAI codecs, and the BD analysis (showing Cheng AT generally saves bits over the AE) are strengths. However, the central quantitative claim is overstated: the paper's own Table 3 does not support 'at least three orders of magnitude ... without hindering' on the Blastocyst dataset, and the DT accounting used for the comparison with Ayad et al. is not sufficiently specified to verify the reported ratios. These issues affect the main advertised conclusion and require correction before publication.

major comments (3)
  1. [Conclusion; Table 3] The conclusion states that SplitFedZip achieves 'transferred data reduction, at least three orders of magnitude compared to no compression, on two medical image segmentation datasets, without hindering the global model performance.' This is not supported by Table 3 for the Blastocyst dataset. With NC MJI = 0.892 and NC DT = 40.6 GB, the best two-stage operating point near NC accuracy is λ = 64 with MJI = 0.884 and DT = 0.4 GB, which is only a 40.6/0.4 ≈ 101x reduction. The rows with CR ≥ 1000 (e.g., λ = 1: MJI 0.810, DT 0.03 GB; λ = 0.6: MJI 0.799, DT 0.02 GB) have MJI well below 0.892, and the two-phase Blastocyst results never reach 0.892 (max MJI 0.847). The conclusion should be revised to state dataset- and operating-point-specific reductions, e.g., roughly 1e2x on Blastocyst and 1e3x or more on HAM10K for near-NC accuracy, or must be backed by a Blastocyst CR curve showing a ≥1e3x point with MJI equal to NC.
  2. [Comparison with existing SplitFed compression methods; Table 3] The data-transfer (DT) figures in Table 3 are not verifiable from the formulas given. For SplitFedZip, DT is defined as 'BPP-T × input image resolution × no. of training samples × 2', but training uses 12 local epochs and 10 global epochs, and no epoch multiplier appears in the formula; the same omission applies to the Ayad et al. DT definition. Moreover, NC DT is listed as 40.6 GB for both Blastocyst (about 664 training samples) and HAM10K (about 8513 training samples). If DT scales with the number of training samples, these two values cannot both be 40.6 GB unless the per-sample uncompressed transfer differs by a factor of roughly 12.8, which is not explained anywhere. Please specify the exact computation, including how many epochs, which image resolution is used for each dataset, and whether forward and backward transmissions are both counted, and report the resulting per-sample and per-epoch DT values so the claimed 3.2x–5000x reductions over Ayad et al. can be checked.
  3. [Results and analysis; Fig. 4] The generalization of the 'at least three orders of magnitude' claim from Fig. 4 to both datasets is not justified in the text. Fig. 4 shows CR versus λ only for the HAM10K dataset, and the passage immediately after it ('In these experiments, SplitFedZip reduces the transferred data by at least three orders of magnitude without compromising model performance') is made in the context of HAM10K examples, yet the conclusion applies it to both datasets. Please either include the corresponding Blastocyst CR curves or restrict the claim to HAM10K and state the Blastocyst reduction separately.
minor comments (5)
  1. [Introduction] The phrase 'where data privacy is of utmost performance' should read 'of utmost importance'.
  2. [Training loss] Equations (1)–(3) are not numbered in the text; please add equation numbers for easier reference, and define the superscripts (e.g., L^{S_i,F}_r) explicitly as 'rate loss at split point i for features'.
  3. [Experimental setup] The sentence 'The horizontal green line represents the MJI during NC (0.892 for both datasets)' should clarify that this is the NC MJI for each dataset separately; both happening to be 0.892 is a coincidence that should be stated explicitly.
  4. [Comparison with existing SplitFed compression methods] The description of the threshold gthres in Ayad et al.'s method is confusing: 'gradients are backpropagated without further compression when the loss exceeds an adaptive threshold (gthres; 0 ≤ gthres ≤ ∞)' should state clearly whether gthres = ∞ means gradients are always compressed, and how the threshold is adapted during training.
  5. [Table 2 caption] The caption 'The average MJI for the complete test sets are listed down' is ungrammatical; please rephrase, e.g., 'The average test-set MJI for each setting is listed below the sample images.'

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the reported rate–accuracy results are empirical sweeps over λ with bitstreams measured from actual codecs, not predictions derived from the fitted inputs.

full rationale

The paper's central result is empirical and self-contained: SplitFedZip is trained with a standard rate–distortion objective (Eq. 1–3), λ is swept over a range, BPP is measured from the actual compressed bitstreams, and MJI is measured on held-out test sets. No parameter is fitted to the reported metric and then relabeled as a prediction; the claimed data-transfer reduction is computed directly from measured bit-rates and training-set sizes. The comparison with Ayad et al. is an external reimplementation of a prior baseline, not a self-referential construction. The self-citations in the paper (split U-Net architecture from Shiranthika et al. 2023 and related SplitFed work) supply architectural context and motivation, but they do not carry the load of the compression claim; the quantitative results would stand identically if those citations were replaced by any other split U-Net implementation. The skeptical concern about the Blastocyst 'at least three orders of magnitude without hindering' claim is an internal-consistency and verifiability issue about the reported table values, not a circularity issue: even if the conclusion overstates the data, the claimed reduction is not equivalent by construction to any fitted input or self-citation chain. Therefore no circular step is present, and the appropriate score is 0.

Assumptions & free parameters 1 free parameters · 4 assumptions · 0 invented entities

No new physical entities or mediators are introduced. The central claim rests on domain assumptions about codec adaptability, the validity of the learned rate estimate, federated averaging of codecs, and the chosen RD loss weighting.

free parameters (1)
  • lambda (rate-distortion trade-off weight) = 2e-9 to 1e10 (swept)
    Controls the balance between bit-rate and distortion/task loss; every R-A curve point corresponds to a different lambda, so the accuracy/compression tradeoff is selected rather than predicted.
assumptions (4)
  • domain assumption Learned image codecs (Balle AE and Cheng AT) can be re-trained to compress feature and gradient tensors of a U-Net without losing task-critical information.
    The whole method assumes the codec can adapt to the statistics of intermediate activations and gradients; no theoretical guarantee is given.
  • domain assumption The entropy-model rate term Lr used in training is an adequate proxy for the actual bitstream length used to compute BPP.
    Training uses Lr from the entropy estimator; reported BPP is actual bits, so any mismatch affects the rate-accuracy tradeoff.
  • domain assumption Federated averaging of codec parameters produces a single global codec that works across heterogeneous client data distributions.
    Used in the aggregation step; no convergence analysis is provided.
  • ad hoc to paper Rate-distortion Lagrangian with a single scalar lambda and additive MSE+Dice distortion is a valid optimization target for the communication-accuracy tradeoff.
    Equations (1)-(3) define the training loss; this weighting is a design choice, not derived.

how reviews work

0 comments
Cite this review

Pith. "Pith review of SplitFedZip: Learned Compression for Data Transfer Reduction in Split-Federated Learning." pith.science (2026). https://pith.science/paper/NVEW24U4

@misc{pith2026241217150,
  author       = {Pith},
  title        = {Pith review of: SplitFedZip: Learned Compression for Data Transfer Reduction in Split-Federated Learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NVEW24U4}},
  note         = {Machine review of arXiv:2412.17150}
}
read the original abstract

Federated Learning (FL) enables multiple clients to train a collaborative model without sharing their local data. Split Learning (SL) allows a model to be trained in a split manner across different locations. Split-Federated (SplitFed) learning is a more recent approach that combines the strengths of FL and SL. SplitFed minimizes the computational burden of FL by balancing computation across clients and servers, while still preserving data privacy. This makes it an ideal learning framework across various domains, especially in healthcare, where data privacy is of utmost importance. However, SplitFed networks encounter numerous communication challenges, such as latency, bandwidth constraints, synchronization overhead, and a large amount of data that needs to be transferred during the learning process. In this paper, we propose SplitFedZip -- a novel method that employs learned compression to reduce data transfer in SplitFed learning. Through experiments on medical image segmentation, we show that learned compression can provide a significant data communication reduction in SplitFed learning, while maintaining the accuracy of the final trained model. The implementation is available at: \url{https://github.com/ChamaniS/SplitFedZip}.

Figures

Figures reproduced from arXiv: 2412.17150 by the authors.

Figure 1
Figure 1. Proposed SplitFedZip architecture. gari et al. 2023) for medical image segmentation. The split U-Net (bottom right corner of [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. R-A curves for the Blastocyst dataset [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. R-A curves for the HAM10K dataset. the goal of compression. Weighted averages of BD-MJI and BD-BPP (WAvg) based on the test dataset sizes show that Cheng AT achieves 68.704% bit-rate savings compared to the AE in the F scheme and 24.029% in the FG scheme (bolded in [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: CR vs. λ for the HAM10K dataset. transferred data by at least three orders of magnitude without compromising model performance. Next, we examine the codec behavior as it adjusts to the statistics of the features and gradients while the main U-Net model is being trained…
Figure 5
Figure 5. Figure 5: Analysis of BPP during SplitFed training for the [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Analysis of BPP during SplitFed training for the [PITH_FULL_IMAGE:figures/full_fig_p006_6.png]

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. AutoEncoder-Compressed Parallel Split Learning for Pre-trained Model Fine-Tuning

    cs.DC 2026-07 conditional novelty 6.0 of 10

    An autoencoder-based split-learning compressor with a two-stage alignment protocol achieves about 10x communication reduction during pre-trained vision-model fine-tuning with near-zero accuracy loss, outperforming heu...

Reference graph

Works this paper leans on

32 extracted references · 25 canonical work pages · cited by 1 Pith paper

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    Ayad, A.; Renner, M.; and Schmeink, A. 2021. Improving the communication and computation efficiency of split learning for iot applications. In Proc. IEEE GLOBECOM, 01--06

    work page 2021

  4. [4]

    Ball \'e , J.; Laparra, V.; and Simoncelli, E. P. 2016. Density modeling of images using a generalized normalization transformation. In Proc. ICLR

    work page 2016

  5. [5]

    J.; and Johnston, N

    Ball \'e , J.; Minnen, D.; Singh, S.; Hwang, S. J.; and Johnston, N. 2018. Variational image compression with a scale hyperprior. In Proc. ICLR

    work page 2018

  6. [6]

    B \'e gaint, J.; Racap \'e , F.; Feltman, S.; and Pushparaja, A. 2020. CompressAI: A PyTorch library and evaluation platform for end-to-end compression research. arXiv preprint arXiv:2011.03029

    arXiv 2020

  7. [7]

    Bjontegaard, G. 2001. Calculation of average PSNR differences between RD-curves. ITU-T SG16 Q, 6

    work page 2001

  8. [8]

    Cheng, Z.; Sun, H.; Takeuchi, M.; and Katto, J. 2020. Learned image compression with discretized gaussian mixture likelihoods and attention modules. In Proc. CVPR, 7939--7948

    work page 2020

Show all 32 references
  1. [9]

    Choi, H.; and Bajić, I. V. 2022. Scalable Video Coding for Humans and Machines. In Proc. IEEE MMSP, 1--6

    2022

  2. [10]

    Cox, M. A. A.; and Cox, T. F. 2008. Multidimensional Scaling, 315--347. Berlin, Heidelberg: Springer Berlin Heidelberg. ISBN 978-3-540-33037-0

    2008

  3. [11]

    Gupta, O.; and Raskar, R. 2018. Distributed learning of deep neural network over multiple agents. J. Netw. Comput., 116: 1--8

    2018

  4. [12]

    Y.; Chuang, Y.-C.; and Wu, A.-Y

    Hsieh, C. Y.; Chuang, Y.-C.; and Wu, A.-Y. 2022. C3 - SL : Circular convolution-based batch-wise compression for communication-efficient split learning. In Proc. IEEE MLSP, 1--6

    2022

  5. [13]

    Isik, B.; Pase, F.; Gunduz, D.; Weissman, T.; and Zorzi, M. 2023. Sparse random networks for communication-efficient federated learning. In Proc. ICLR

    2023

  6. [14]

    H.; Shiranthika, C.; Saeedi, P.; and Bajić, I

    Kafshgari, Z. H.; Shiranthika, C.; Saeedi, P.; and Bajić, I. V. 2023. Quality-Adaptive Split-Federated Learning for Segmenting Medical Images with Inaccurate Annotations. In Proc. IEEE ISBI, 1--5

    2023

  7. [15]

    B.; Yu, F

    Kone c n \`y , J.; McMahan, H. B.; Yu, F. X.; Richt \'a rik, P.; Suresh, A. T.; and Bacon, D. 2016. Federated learning: Strategies for improving communication efficiency. In Proc. NeurIPS, 5--10

    2016

  8. [16]

    Kong, Y.; Yu, W.; Xu, S.; Yu, F.; Xu, Y.; and Huang, Y. 2024. Two Birds With One Stone: Towards Communication and Computation Efficient Federated Learning. IEEE Commun. Lett

    2024

  9. [17]

    Lockhart, L.; Saeedi, P.; Au, J.; and Havelock, J. 2019. Multi-Label Classification for Automatic Human Blastocyst Grading with Severely Imbalanced Data. In Proc. IEEE MMSP, 1--6

    2019

  10. [18]

    J.; Malekijou, H.; Homayounfar, M.; Alizadeh-Shabdiz, F.; and Rawassizadeh, R

    Malekijoo, A.; Fadaeieslam, M. J.; Malekijou, H.; Homayounfar, M.; Alizadeh-Shabdiz, F.; and Rawassizadeh, R. 2021. Fedzip: A compression framework for communication-efficient federated learning. arXiv preprint arXiv:2102.01593

    2021 arXiv

  11. [19]

    McMahan, B.; Moore, E.; Ramage, D.; Hampson, S.; and y Arcas, B. A. 2017. Communication - Efficient Learning of Deep Networks from Decentralized Data . In Proc. AISTATS, 1273--1282. PMLR

    2017

  12. [20]

    Mitchell, N.; Ball \'e , J.; Charles, Z.; and Kone c n \`y , J. 2022. Optimizing the communication-accuracy trade-off in federated learning with rate-distortion theory. arXiv preprint arXiv:2201.02664

    2022 arXiv

  13. [21]

    Mudvari, A.; Vainio, A.; Ofeidis, I.; Tarkoma, S.; and Tassiulas, L. 2024 a . Adaptive compression-aware split learning and inference for enhanced network efficiency. ACM TOIT, 24(4): 1--26

    2024

  14. [22]

    Mudvari, A.; Vainio, A.; Ofeidis, I.; Tarkoma, S.; and Tassiulas, L. 2024 b . Adaptive Compression-Aware Split Learning and Inference for Enhanced Network Efficiency. arXiv preprint arXiv:2311.05739

    2024 arXiv

  15. [23]

    G.; and Jeon, Y.-S

    Oh, Y.; Lee, J.; Brinton, C. G.; and Jeon, Y.-S. 2023. Communication-Efficient Split Learning via Adaptive Feature-Wise Compression. arXiv preprint arXiv:2307.10805

    2023 arXiv

  16. [24]

    H.; Saeedi, P.; and Baji \'c , I

    Shiranthika, C.; Kafshgari, Z. H.; Saeedi, P.; and Baji \'c , I. V. 2023. SplitFed resilience to packet loss: Where to split, that is the question. In Proc. MICAAI, 367--377. Springer

    2023

  17. [25]

    Shiranthika, C.; Saeedi, P.; and Bajić, I. V. 2023. Decentralized Learning in Healthcare: A Review of Emerging Techniques. IEEE Access, 11: 54188--54209

    2023

  18. [26]

    H.; Li, W.; Vercauteren, T.; Ourselin, S.; and Jorge Cardoso, M

    Sudre, C. H.; Li, W.; Vercauteren, T.; Ourselin, S.; and Jorge Cardoso, M. 2017. Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations. In DLMIA ML-CDS, 240--248. Springer

    2017

  19. [27]

    Thapa, C.; Arachchige, P. C. M.; Camtepe, S.; and Sun, L. 2022. SplitFed : When Federated Learning Meets Split Learning . In Proc. AAAI, volume 36, 8485--8493

    2022

  20. [28]

    Tschandl, P.; Rosendahl, C.; and Kittler, H. 2018. The HAM10000 dataset, a large collection of multi-source dermatoscopic images of common pigmented skin lesions. Sci. Data, 5(1): 1--9

    2018

  21. [29]

    S.; Reddi, S.; Waghmare, S.; Yu, F

    Wang, J.; Qi, H.; Rawat, A. S.; Reddi, S.; Waghmare, S.; Yu, F. X.; and Joshi, G. 2022. Fedlite: A scalable approach for federated learning on resource-constrained clients. arXiv preprint arXiv:2201.11865

    2022 arXiv

  22. [30]

    Wu, C.; Wu, F.; Lyu, L.; Huang, Y.; and Xie, X. 2022. Communication-efficient federated learning via knowledge distillation. Nat. Commun., 13(1): 2032

    2022

  23. [31]

    H.; Amer, H.; and Jiang, Y

    Yang, E. H.; Amer, H.; and Jiang, Y. 2021. Compression Helps Deep Learning in Image Classification. Entropy, 23(7)

    2021

  24. [32]

    Zheng, F.; Chen, C.; Lyu, L.; and Yao, B. 2023. Reducing communication for split learning by randomized top-k sparsification. In Proc. IJCAI, 4665--4673

    2023

Pith tools

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