Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

FedAli: Personalized Federated Learning Alignment with Prototype Layers for Generalized Mobile Services

T0 review · 4 major / 5 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read FedAli claims that an alignment layer matching embeddings to dual prototype banks lets personalized federated learning generalize across mobile clients without sacrificing local accuracy.

desk verdict Novel prototype layer and clear write-up, but the headline generalization claim overreaches the metric, which never holds out a client or domain. read the letter →

arxiv 2411.10595 v2 pith:QVA6IHFI submitted 2024-11-15 cs.LG cs.CRcs.CV

Sannara Ek , Kaile Wang , François Portet , Philippe Lalanda , Jiannong Cao This is my paper · ORCID
classification cs.LGcs.CRcs.CV
keywords personalizedfederatedlearningprototypealignmentoptimaltransportSinkhorn-Knoppnon-IIDdatahumanactivityrecognitionmobilesensingdomaingeneralization
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

FedAli is a personalized federated learning strategy aimed at a property most PFL methods do not evaluate: a client model should stay accurate when it meets data from other clients' environments, not just on its own local data. The paper argues that existing prototype-based methods fail on this because they tie a single global prototype to each class label and never adapt the model's embeddings at inference time. FedAli instead inserts an alignment layer into each transformer block that matches embeddings to two prototype banks—local ones updated per client, global ones formed by clustering across clients—using a label-free optimal transport plan. The reported experiments on wearable human-activity-recognition and CIFAR benchmarks claim that this dual prototype alignment raises cross-client generalization substantially while keeping personalization at or near the best baseline. If that is right, FedAli is a drop-in layer that improves generalization in personalized mobile sensing without sacrificing the personal fit.

What carries the argument

The central object is the ALignment with Prototypes (ALP) layer, placed before the multi-head attention in every encoder block of a small transformer. The layer receives a sequence of embeddings and computes a Sinkhorn-Knopp optimal transport plan between those embeddings and two prototype banks: local prototypes, updated by exponential moving average and acting as the client's memory, and global prototypes, produced on the server by K-means clustering of all clients' local prototypes each round. During training the transport plan assigns embeddings to global prototypes, the matched prototypes pass through a gated linear unit, and are blended into the embeddings with coefficient $\beta$. During inference only local prototypes are matched, pulling unseen inputs toward the client's learned geometry. This label-free matching is what lets the layer redirect embeddings toward a shared representation without a per-class prototype.

What would settle it

Train FedAli with an entire subject, device model, or body position held out from all clients, then evaluate on that held-out domain; if the cross-client generalization score does not improve over FedAvg when the test distribution was never present in training, the central claim fails.

Watch

Extended reading notes

Core claim

FedAli's central claim is that a personalized federated model can be made more transferable by aligning its embeddings to prototypes at two different moments: during training, shared global prototypes pull all clients' representations into a common coordinate system, and during inference, client-specific local prototypes pull novel inputs toward the geometry the client learned. The paper packages this as the ALignment with Prototypes (ALP) layer, inserted at the start of each transformer encoder block. The layer computes a Sinkhorn-Knopp optimal transport plan between the input embeddings and the two prototype banks, applies a gated linear unit to the matched prototypes, and blends them into the embeddings with a weight $\beta$. Because the prototypes are feature-based rather than class-based, no labels are required, and the whole system can be pre-trained with a masked autoencoder. Empirically, the paper reports that on HHAR the generalization score rises from 75.85 percent for FedAvg to 81.84 percent while personalization rises from 96.88 to 98.15 percent, and on CIFAR-100 with 20 clients it reports the best personalization and generalization among the compared strategies.

Load-bearing premise

The load-bearing premise is that each client model's averaged accuracy on the test sets of all participating clients is a valid proxy for performance in genuinely unseen mobile environments; if that proxy is weak, the claimed generalization gains are not established.

Editorial extensions

If this is right

  • FedAli can be added to an existing transformer model and trained under standard FedAvg-style aggregation, with prototypes transmitted alongside weights, so the method is a layer-level upgrade rather than a new optimization framework.
  • Because prototype alignment needs no class labels, the same ALP layer can be pre-trained in a self-supervised way on unlabeled sensor streams; the paper reports that this pre-training cuts convergence time and yields more than a 20 percent generalization gain on HHAR.
  • Client models trained with FedAli should transfer better across users, devices, and on-body positions, since the t-SNE evidence shows same activities from different datasets mapped into a shared region of embedding space.
  • The personalization-generalization trade-off is relaxed: global prototypes anchor training embeddings across clients while local prototypes anchor inference to the client's own distribution, so a client need not sacrifice local accuracy to gain cross-client generalization.
  • The extra cost is bounded: prototypes are non-trainable, inference uses only local prototypes, and local training time is 1.09 times FedAvg, faster than contrastive baselines such as MOON and FedPAC, though communication overhead is 1.84 times FedAvg.

Reading between the lines

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

  • A true test of the generalization claim would hold out an entire subject, device, or body position from all clients and measure accuracy on that never-seen domain; the paper's benchmark averages over test sets of participating clients, which may overstate transfer to genuinely novel environments.
  • The label-free prototype matching could be adapted to federated regression, forecasting, or anomaly detection, where class labels are not well defined; whether the Sinkhorn-Knopp plan remains stable with continuous targets is an open question.
  • Since global prototypes are formed by K-means over clients' local prototypes each round, the method's behavior under client churn and partial participation has not been characterized; a natural extension is to study convergence when only a subset of clients uploads prototypes.
  • The ALP layer's position before attention suggests a design space: placing it at different depths or using different prototype counts per layer may trade off personalization and generalization in ways the sensitivity analysis only begins to explore.
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 / 5 minor

Summary. The paper introduces FedAli, a personalized federated learning strategy that adds an ALignment with Prototypes (ALP) layer to transformer-based models. The layer matches input embeddings to local and global prototype banks via Sinkhorn optimal transport, aligns embeddings with the matched prototypes through a gated linear unit, and updates local prototypes by EMA during training; the server aggregates prototypes by K-means clustering. The method is evaluated on three HAR datasets and two vision datasets, reporting personalization, generalization, and global scores, plus a pre-trained initialization study and ablation and sensitivity analyses. The abstract claims that FedAli 'significantly enhances client generalization while preserving strong personalization in heterogeneous settings.'

Significance. The proposed label-free prototype layer is a plausible and relatively lightweight mechanism for reducing representation drift in federated learning, and the paper has tangible strengths: the source code is released, the ablation study isolates the contribution of each component, the communication and computation overhead is reported, and the pre-training study addresses a practically relevant question. If the central claim were established, FedAli would be a useful drop-in component for mobile FL. However, as reported, the evidence supports only modest and inconsistent gains, and the generalization metric does not measure the 'unseen scenarios' that motivate the work. The contribution is therefore promising but not yet convincingly validated.

major comments (4)
  1. [Section 5.1.4 and Section 1] The 'Generalization score' is defined as the average performance of each client model on the test sets of all participating clients. This metric does not measure robustness to the unseen scenarios that the introduction motivates: a new user, a new phone model, or a new body position. No client, device, or on-body position is held out from training, and the t-SNE 'unseen' client in Section 5.3 is still a participant in the Combined federation. The central claim therefore needs either a leave-one-client-out or leave-one-device/position-out evaluation, or a revised claim that is explicitly limited to inter-client transfer among participating clients.
  2. [Tables 1 and 2] The claimed significant generalization advantage is not consistently present across the six experimental settings. Concretely, the generalization gain over FedAvg is about +6.0 points on HHAR, but only +0.9 on RealWorld and +1.0 on Combined; on CIFAR10 with 20 clients FedAli is slightly behind FedAvg (43.14 vs. 43.29), and on CIFAR100 with 50 clients the gain is +0.9 points. The global score on RealWorld is below FedAvg (71.50 vs. 72.47). Only standard deviations over clients are reported, with no statistical significance tests or confidence intervals over repeated runs, so the abstract's 'significantly enhances' is not supported by the tables as presented.
  3. [Section 6.1, Tables 5 and 6] The hyperparameters beta = 0.2 and the prototype counts were selected on the HHAR dataset using the same generalization metric that is later reported as the headline result in Table 1. This selection on the evaluation dataset can inflate the apparent HHAR advantage, and no validation protocol is described for the other datasets. Additionally, Table 5 shows that beta = 0.5 yields a higher generalization score (70.30 vs. 70.04) and the same global score (83.00) as beta = 0.2, so the claim that beta = 0.2 gives the 'best balance' is not clearly supported by the table.
  4. [Section 5.4 and Figures 6 and 7] The pre-training section claims large gains ('more than 20%' in generalization and 'more than 11%' in global performance) and faster convergence, but these claims are made without a quantitative table or error bars, relying on learning curves that are not numerically summarized. For a load-bearing performance claim about convergence and pre-training benefits, the authors should report final numbers with run-level variability for all compared strategies.
minor comments (5)
  1. [Section 5.1.2] The text says 20 and 100 clients are generated for the vision datasets, but Table 2 reports experiments with 20 and 50 clients; the text and the table need to be reconciled.
  2. [Section 5.1.4] The personalization score is described as a macro F1 score, while the generalization and global scores are described as 'average performance' or 'accuracy'; the manuscript should state explicitly whether all three metrics use the same measure and how the macro averaging is performed.
  3. [Section 2, Equation (2)] The displayed PFL objective is not mathematically well-formed: it defines F as a sum of local losses and then introduces '∀c min w_c f_c(w_c; D_c)' with a constraint on H(w_c; D), but the relationship between F, the inner minimization, and the constraint is unclear.
  4. [Section 6] The ablation and sensitivity tables report plus-minus values from 'three random runs', but it is not stated whether the spread is over runs or over clients; this should be clarified because client-level heterogeneity is large in these datasets.
  5. [Section 5.3] The phrase 'unseen RealWorld client' is misleading: the client participates in the federated training process and is only unseen to the particular HHAR local model. The authors should qualify this as 'not locally seen' to avoid overstating the out-of-distribution nature of the experiment.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: FedAli is an empirical architecture paper with no derivation chain that reduces its claims to its inputs.

full rationale

FedAli does not derive a prediction from fitted quantities by construction. The ALP layer is defined operationally (Algorithm 1: Sinkhorn optimal transport, GLU gating, and the convex combination x_hat = beta*P_GLU + (1-beta)*x), and FedAli aggregation is algorithmic (K-means over local prototypes). There is no equation-level equivalence between the claimed generalization improvements and these definitions. The only self-citation is the HART backbone ([15]), which is a design choice shared by all compared methods and is therefore not load-bearing. The closest evaluation concern is Section 6.1, where beta and prototype counts are tuned on HHAR before the HHAR results in Table 1 are reported; this is a model-selection and benchmark-reuse issue, not circularity, because the 600-round results are empirical outcomes rather than the tuning objective recomputed, and the paper also reports partially independent results on RealWorld, Combined, CIFAR-10, and CIFAR-100. Section 5.1.4 defines Generalization as average performance on all participating clients' test sets, which limits external-validity claims about truly novel users or devices, but does not make the method's output equal to that metric. The paper even records a concrete failure mode in Section 5.3, noting that the Walking activity remains misaligned because of its similarity to Upstair, which confirms the alignment mechanism has real content and is not tautological.

Assumptions & free parameters 6 free parameters · 6 assumptions · 1 invented entities

The method is empirical: there is no mathematical derivation of why prototype alignment improves generalization. Key coefficients beta and gamma, prototype counts, and KMeans cluster counts are hand-tuned or inherited from prior work. The load-bearing hypotheses are the quality of the OT assignment, the EMA update rule, the convex embedding alignment rule, and the validity of the generalization metric.

free parameters (6)
  • Alignment coefficient beta = 0.2
    Controls prototype influence in Eq. (4); tuned on HHAR in Section 6.1; lower values favor personalization, higher values favor generalization.
  • Prototype EMA update rate gamma = 0.999
    Used in Eq. (5) for local prototype updates; set in Section 5.1.3 after tuning, but no sensitivity experiment is reported.
  • Prototype counts per encoder block G = 2048, 1024, 512, 256, 128, 64
    Selected from Table 6 on HHAR; larger counts add parameters with marginal gains.
  • Sinkhorn sharpness epsilon = 0.05
    Inherited from [7], not tuned in this paper; affects the sharpness of the optimal transport plan.
  • Sinkhorn normalization iterations = 3
    Inherited from [7], not tuned; balances performance and efficiency.
  • KMeans cluster count = G per layer
    Set equal to the prototype count in Algorithm 2; no independent tuning is described.
assumptions (6)
  • domain assumption Sinkhorn-Knopp with 3 iterations yields a transport plan accurate enough for prototype assignment (Eq. 3).
    The ALP layer relies on the quality of the OT plan; no convergence analysis is given for 3 iterations in this setting.
  • domain assumption EMA updates of local prototypes with gamma = 0.999 let prototypes track local data statistics without gradient supervision.
    Eq. (5); no analysis of EMA bias or variance is provided.
  • ad hoc to paper Convex interpolation in Eq. (4) pulls embeddings toward prototypes and improves generalization without destroying personalization.
    This is the core design hypothesis of the paper; there is no derivation, only ablations on HHAR.
  • domain assumption KMeans centroids of local prototypes define stable shared global anchors across communication rounds.
    Algorithm 2; cluster index permutation and round-to-round stability are not analyzed.
  • domain assumption Average test accuracy across other clients' test sets measures generalization to unseen environments.
    Section 5.1.4; test sets come from the same subject and device distributions as other clients' training data, not from genuinely novel environments.
  • domain assumption MAE pre-training on five public HAR datasets transfers to HHAR, RealWorld, Combined, and to the vision backbones.
    Section 5.4; a standard transfer assumption, evaluated only for HAR and not for the vision experiments.
invented entities (1)
  • Local and global prototype banks (ALP layer memories)
    purpose: Act as learned anchors that align embeddings at inference (local prototypes) and during federated training (global prototypes).
    These are learned parameters inside the model; the paper offers no externally falsifiable prediction about them, so they function as internal constructs rather than independently evidenced entities.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FedAli: Personalized Federated Learning Alignment with Prototype Layers for Generalized Mobile Services." pith.science (2026). https://pith.science/paper/QVA6IHFI

@misc{pith2026241110595,
  author       = {Pith},
  title        = {Pith review of: FedAli: Personalized Federated Learning Alignment with Prototype Layers for Generalized Mobile Services},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/QVA6IHFI}},
  note         = {Machine review of arXiv:2411.10595}
}
read the original abstract

Personalized Federated Learning (PFL) enables distributed training on edge devices, allowing models to collaboratively learn global patterns while tailoring their parameters to better fit each client's local data, all while preserving data privacy. However, PFL faces two key challenges in mobile systems: client drift, where heterogeneous data cause model divergence, and the overlooked need for client generalization, as the dynamic of mobile sensing demands adaptation beyond local environments. To overcome these limitations, we introduce Federated Alignment (FedAli), a prototype-based regularization technique that enhances inter-client alignment while strengthening the robustness of personalized adaptations. At its core, FedAli introduces the ALignment with Prototypes (ALP) layer, inspired by human memory, to enhance generalization by guiding inference embeddings toward personalized prototypes while reducing client drift through alignment with shared prototypes during training. By leveraging an optimal transport plan to compute prototype-embedding assignments, our approach allows pre-training the prototypes without any class labels to further accelerate convergence and improve performance. Our extensive experiments show that FedAli significantly enhances client generalization while preserving strong personalization in heterogeneous settings.

Figures

Figures reproduced from arXiv: 2411.10595 by the authors.

Figure 1
Figure 1. Federated Alignment with heterogeneous clients [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. The ALP layer is positioned after the first normalization layer within the transformer encoder [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. The five processes of the ALP layer during training mode [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: t-SNE visualization comparing FedAvg (left) and FedAli (right). The inference representation of [PITH_FULL_IMAGE:figures/full_fig_p013_4.png]
Figure 5
Figure 5. Figure 5: t-SNE visualization comparing FedAvg (left) and FedAli (right). The training data representations [PITH_FULL_IMAGE:figures/full_fig_p014_5.png]
Figure 6
Figure 6. Figure 6: Learning curve of comparison of starting from randomly initialized weights (Scratch) versus starting [PITH_FULL_IMAGE:figures/full_fig_p015_6.png]
Figure 8
Figure 8. Figure 8: Visualization of parameter distribution among the Base model, GLU, Local Prototype, and Global [PITH_FULL_IMAGE:figures/full_fig_p015_8.png]
Figure 7
Figure 7. Figure 7: Learning curves of FL strategies on the HHAR, RealWorld and the Combined datasets. [PITH_FULL_IMAGE:figures/full_fig_p016_7.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. FedDOSE: Federated Learning Framework Decomposing Site Effects for Modeling Brain Dynamic Functional Connectivity

    cs.LG 2026-08 conditional novelty 6.0 of 10

    A federated framework that decomposes multi-site brain dynamic connectivity into disease, phenotype, and scanner parts and aligns site prototypes to improve ASD and ADHD classification.

Reference graph

Works this paper leans on

68 extracted references · 48 canonical work pages · cited by 1 Pith paper

  1. [1]

    Durmus Alp Emre Acar, Yue Zhao, Ruizhao Zhu, Ramon Matas, Matthew Mattina, Paul Whatmough, and Venkatesh Saligrama. 2021. Debiasing model updates for improving personalized federated training. In International conference on machine learning. PMLR, 21–31

    work page 2021

  2. [2]

    Davide Anguita, Alessandro Ghio, Luca Oneto, Xavier Parra, and Jorge Luis Reyes-Ortiz. 2013. A Pub- lic Domain Dataset for Human Activity Recognition using Smartphones. In 21st European Symposium on Artificial Neural Networks, ESANN 2013, Bruges, Belgium, April 24-26, 2013. 17

    work page 2013

  3. [3]

    Manoj Ghuhan Arivazhagan, Vinay Aggarwal, Aaditya Kumar Singh, and Sunav Choudhary. 2019. Federated learning with personalization layers. arXiv preprint arXiv:1912.00818 (2019)

    arXiv 2019

  4. [4]

    Lawrence W Barsalou. 2009. Simulation, situated conceptualization, and prediction. Philosophical transactions of The Royal Society B: biological sciences 364, 1521 (2009), 1281–1289

    work page 2009

  5. [5]

    Dongqi Cai, Shangguang Wang, Yaozong Wu, Felix Xiaozhu Lin, and Mengwei Xu. 2023. Federated few-shot learning for mobile nlp. In Proceedings of the 29th Annual International Conference on Mobile Computing and Networking. 1–17

    work page 2023

  6. [6]

    Dongqi Cai, Yaozong Wu, Shangguang Wang, Felix Xiaozhu Lin, and Mengwei Xu. 2023. Efficient Federated Learning for Modern NLP. In Proceedings of the 29th Annual International Conference on Mobile Computing and Networking (Madrid, Spain) (ACM MobiCom ’23). Association for Computing Machinery, New York, NY, USA, Article 37, 16 pages. https://doi.org/10.1145...

    arXiv 2023

  7. [7]

    Mathilde Caron, Ishan Misra, Julien Mairal, Priya Goyal, Piotr Bojanowski, and Armand Joulin

    work page

  8. [8]

    Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. 2020. A simple framework for contrastive learning of visual representations. In International conference on machine learning. PMLR, 1597–1607

    work page 2020

Show all 68 references
  1. [9]

    Yiqiang Chen, Xin Qin, Jindong Wang, Chaohui Yu, and Wen Gao. 2020. Fedhealth: A federated transfer learning framework for wearable healthcare. IEEE Intelligent Systems 35, 4 (2020), 83–93

    2020

  2. [10]

    Zihan Chen, Howard Hao Yang, Tony Quek, and Kai Fong Ernest Chong. 2023. Spectral Co-Distillation for Personalized Federated Learning. In Thirty-seventh Conference on Neural Information Processing Systems

    2023

  3. [11]

    Liam Collins, Hamed Hassani, Aryan Mokhtari, and Sanjay Shakkottai. 2021. Exploiting shared repre- sentations for personalized federated learning. In International conference on machine learning. PMLR, 2089–2099

    2021

  4. [12]

    Marco Cuturi. 2013. Sinkhorn distances: Lightspeed computation of optimal transport. Advances in neural information processing systems 26 (2013)

    2013

  5. [13]

    Yutong Dai, Zeyuan Chen, Junnan Li, Shelby Heinecke, Lichao Sun, and Ran Xu. 2023. Tackling data heterogeneity in federated learning with class prototypes. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 37. 7314–7322

    2023

  6. [14]

    Yann N Dauphin, Angela Fan, Michael Auli, and David Grangier. 2017. Language modeling with gated convolutional networks. In International conference on machine learning. PMLR, 933–941

    2017

  7. [15]

    Sannara Ek, Fran¸ cois Portet, and Philippe Lalanda. 2023. Transformer-based models to deal with heterogeneous environments in Human Activity Recognition. Personal and Ubiquitous Computing (2023), 1–14

    2023

  8. [16]

    Hristijan Gjoreski, Mathias Ciliberto, Lin Wang, Francisco Javier Ordonez Morales, Sami Mekki, Stefan Valentin, and Daniel Roggen. 2018. The University of Sussex-Huawei Locomotion and Transportation Dataset for Multimodal Analytics With Mobile Devices. IEEE Access 6 (2018), 42...

    2018

  9. [17]

    Yaming Guo, Kai Guo, Xiaofeng Cao, Tieru Wu, and Yi Chang. 2023. Out-of-distribution generaliza- tion of federated learning via implicit invariant relationships. In International Conference on Machine Learning. PMLR, 11905–11933

    2023

  10. [18]

    Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Doll´ ar, and Ross Girshick. 2022. Masked autoencoders are scalable vision learners. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 16000–16009. 18

    2022

  11. [19]

    Geoffrey Hinton, Oriol Vinyals, and Jeff Dean. 2015. Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531 (2015)

    2015 arXiv

  12. [20]

    Peter Kairouz, H Brendan McMahan, Brendan Avent, Aur´ elien Bellet, Mehdi Bennis, Arjun Nitin Bhagoji, Kallista Bonawitz, Zachary Charles, Graham Cormode, Rachel Cummings, et al. 2021. Ad- vances and open problems in federated learning. Foundations and trends® in machine learn...

    2021

  13. [21]

    Alex Krizhevsky and Geoffrey Hinton. 2009. Learning multiple layers of features from tiny images. Technical Report 0. University of Toronto, Toronto, Ontario

    2009

  14. [22]

    Stefanos Laskaridis, Stylianos I Venieris, Alexandros Kouris, Rui Li, and Nicholas D Lane. 2024. The future of consumer edge-ai computing. IEEE Pervasive Computing (2024)

    2024

  15. [23]

    Daliang Li and Junpu Wang. 2019. Fedmd: Heterogenous federated learning via model distillation. arXiv preprint arXiv:1910.03581 (2019)

    2019 arXiv

  16. [24]

    Hongxia Li, Zhongyi Cai, Jingya Wang, Jiangnan Tang, Weiping Ding, Chin-Teng Lin, and Ye Shi

  17. [25]

    Mohan Li, Martin Gjoreski, Pietro Barbiero, Gaˇ sper Slapniˇ car, Mitja Luˇ strek, Nicholas D Lane, and Marc Langheinrich. 2025. A Survey on Federated Learning in Human Sensing. arXiv preprint arXiv:2501.04000 (2025)

    2025 arXiv

  18. [26]

    Qinbin Li, Bingsheng He, and Dawn Song. 2021. Model-contrastive federated learning. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 10713–10722

    2021

  19. [27]

    Qinbin Li, Zeyi Wen, Zhaomin Wu, Sixu Hu, Naibo Wang, Yuan Li, Xu Liu, and Bingsheng He. 2021. A survey on federated learning systems: Vision, hype and reality for data privacy and protection. IEEE Transactions on Knowledge and Data Engineering 35, 4 (2021), 3347–3366

    2021

  20. [28]

    Tian Li, Shengyuan Hu, Ahmad Beirami, and Virginia Smith. 2021. Ditto: Fair and robust federated learning through personalization. In International conference on machine learning. PMLR, 6357–6368

    2021

  21. [29]

    Tian Li, Anit Kumar Sahu, Ameet Talwalkar, and Virginia Smith. 2020. Federated learning: Challenges, methods, and future directions. IEEE signal processing magazine 37, 3 (2020), 50–60

    2020

  22. [30]

    Tian Li, Anit Kumar Sahu, Manzil Zaheer, Maziar Sanjabi, Ameet Talwalkar, and Virginia Smith

  23. [31]

    Xiang Li, Kaixuan Huang, Wenhao Yang, Shusen Wang, and Zhihua Zhang. 2020. On the Convergence of FedAvg on Non-IID Data. In International Conference on Learning Representations

    2020

  24. [32]

    Paul Pu Liang, Terrance Liu, Liu Ziyin, Nicholas B Allen, Randy P Auerbach, David Brent, Ruslan Salakhutdinov, and Louis-Philippe Morency. 2020. Think locally, act globally: Federated learning with local and global representations. arXiv preprint arXiv:2001.01523 (2020)

    2020 arXiv

  25. [33]

    Proceedings of Machine learning and systems 2 (2020), 429–450

    Federated optimization in heterogeneous networks. Proceedings of Machine learning and systems 2 (2020), 429–450

    2020

  26. [34]

    Stuart Lloyd. 1982. Least squares quantization in PCM. IEEE transactions on information theory 28, 2 (1982), 129–137

    1982

  27. [35]

    Wang Lu, Jindong Wang, Yiqiang Chen, Sinno Jialin Pan, Chunyu Hu, and Xin Qin. 2022. Semantic- discriminative mixup for generalizable sensor-based cross-domain activity recognition. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 6, 2 (2022)...

    2022

  28. [36]

    Yunming Liao, Yang Xu, Hongli Xu, Lun Wang, and Chen Qian. 2023. Adaptive configuration for het- erogeneous participants in decentralized federated learning. In IEEE INFOCOM 2023-IEEE Conference on Computer Communications. IEEE, 1–10

    2023

  29. [37]

    Othmane Marfoq, Giovanni Neglia, Richard Vidal, and Laetitia Kameni. 2022. Personalized federated learning through local memorization. In International Conference on Machine Learning. PMLR, 15070– 15092

    2022

  30. [38]

    Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hampson, and Blaise Aguera y Arcas. 2017. Communication-efficient learning of deep networks from decentralized data. InArtificial intelligence and statistics. PMLR, 1273–1282

    2017

  31. [39]

    Clegg, Andrea Cavallaro, and Hamed Haddadi

    Mohammad Malekzadeh, Richard G. Clegg, Andrea Cavallaro, and Hamed Haddadi. 2018. Protecting Sensory Data Against Sensitive Inferences. In Proceedings of the 1st Workshop on Privacy by Design in Distributed Systems (Porto, Portugal) (W-P2DS’18). ACM, New York, NY, USA, Article...

    2018

  32. [40]

    Xutong Mu, Yulong Shen, Ke Cheng, Xueli Geng, Jiaxuan Fu, Tao Zhang, and Zhiwei Zhang. 2023. Fedproc: Prototypical contrastive federated learning on non-iid data. Future Generation Computer Systems 143 (2023), 93–104

    2023

  33. [41]

    Subhas Chandra Mukhopadhyay. 2014. Wearable sensors for human activity monitoring: A review. IEEE sensors journal 15, 3 (2014), 1321–1330

    2014

  34. [42]

    Umberto Michieli and Mete Ozay. 2021. Prototype guided federated learning of visual feature represen- tations. arXiv preprint arXiv:2105.08982 (2021)

    2021 arXiv

  35. [43]

    John Nguyen, Jianyu Wang, Kshitiz Malik, Maziar Sanjabi, and Michael Rabbat. 2023. Where to Begin? On the Impact of Pre-Training and Initialization in Federated Learning. In The Eleventh International Conference on Learning Representations

    2023

  36. [44]

    Gabriel Peyr´ e, Marco Cuturi, et al. 2019. Computational optimal transport: With applications to data science. Foundations and Trends® in Machine Learning 11, 5-6 (2019), 355–607

    2019

  37. [45]

    Tuan Nguyen, Philip Torr, and Ser-Nam Lim

    A. Tuan Nguyen, Philip Torr, and Ser-Nam Lim. 2022. FedSR: A Simple and Effective Domain Gen- eralization Method for Federated Learning. In Advances in Neural Information Processing Systems, Alice H. Oh, Alekh Agarwal, Danielle Belgrave, and Kyunghyun Cho (Eds.)

    2022

  38. [46]

    Xin Qin, Yiqiang Chen, Jindong Wang, and Chaohui Yu. 2019. Cross-dataset activity recognition via adaptive spatial-temporal transfer learning. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies 3, 4 (2019), 1–25

    2019

  39. [47]

    Lawrence

    Joaquin Quionero-Candela, Masashi Sugiyama, Anton Schwaighofer, and Neil D. Lawrence. 2009. Dataset Shift in Machine Learning. The MIT Press

    2009

  40. [48]

    Thomas Pl¨ otz. 2023. If only we had more data!: sensor-based human activity recognition in challeng- ing scenarios. In 2023 IEEE International Conference on Pervasive Computing and Communications Workshops and other Affiliated Events (PerCom Workshops). IEEE, 565–570

    2023

  41. [49]

    Yujin Shin, Kichang Lee, Sungmin Lee, You Rim Choi, Hyung-Sin Kim, and JeongGil Ko. 2024. Effective Heterogeneous Federated Learning via Efficient Hypernetwork-based Weight Generation. InProceedings of the 22nd ACM Conference on Embedded Networked Sensor Systems. 112–125

    2024

  42. [50]

    Jake Snell, Kevin Swersky, and Richard Zemel. 2017. Prototypical networks for few-shot learning. Advances in neural information processing systems 30 (2017)

    2017

  43. [51]

    Attila Reiss and Didier Stricker. 2012. Introducing a new benchmarked dataset for activity monitoring. In 2012 16th International Symposium on Wearable Computers. IEEE, 108–109

    2012

  44. [52]

    Sztyler and H

    T. Sztyler and H. Stuckenschmidt. 2016. On-body localization of wearable devices: An investigation of position-aware activity recognition. In 2016 IEEE International Conference on Pervasive Computing and Communications (PerCom). 1–9

    2016

  45. [53]

    Alysa Ziying Tan, Han Yu, Lizhen Cui, and Qiang Yang. 2022. Towards personalized federated learning. IEEE Transactions on Neural Networks and Learning Systems (2022)

    2022

  46. [54]

    Allan Stisen, Henrik Blunck, Sourav Bhattacharya, Thor Siiger Prentow, Mikkel Baun Kjærgaard, Anind Dey, Tobias Sonne, and Mads Møller Jensen. 2015. Smart devices are different: Assessing and mitigatingmobile sensing heterogeneities for activity recognition. In Proceedings of ...

    2015

  47. [55]

    Yue Tan, Guodong Long, Jie Ma, Lu Liu, Tianyi Zhou, and Jing Jiang. 2022. Federated learning from pre-trained models: A contrastive learning approach. Advances in neural information processing systems 35 (2022), 19332–19344

    2022

  48. [56]

    Laurens Van der Maaten and Geoffrey Hinton. 2008. Visualizing data using t-SNE. Journal of machine learning research 9, 11 (2008)

    2008

  49. [57]

    Yue Tan, Guodong Long, Lu Liu, Tianyi Zhou, Qinghua Lu, Jing Jiang, and Chengqi Zhang. 2022. Fedproto: Federated prototype learning across heterogeneous clients. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 36. 8432–8440

    2022

  50. [58]

    C´ edric Villani et al. 2009. Optimal transport: old and new. Vol. 338. Springer

    2009

  51. [59]

    Jindong Wang, Cuiling Lan, Chang Liu, Yidong Ouyang, Tao Qin, Wang Lu, Yiqiang Chen, Wenjun Zeng, and S Yu Philip. 2022. Generalizing to unseen domains: A survey on domain generalization. IEEE transactions on knowledge and data engineering 35, 8 (2022), 8052–8072

    2022

  52. [60]

    George Vavoulas, Charikleia Chatzaki, Thodoris Malliotakis, Matthew Pediaditis, and Manolis Tsik- nakis. 2016. The MobiAct Dataset: Recognition of Activities of Daily Living using Smartphones.. In ICT4AgeingWell. 143–151

    2016

  53. [61]

    Jian Xu, Xinyi Tong, and Shao-Lun Huang. 2023. Personalized Federated Learning with Fea- ture Alignment and Classifier Collaboration. In The Eleventh International Conference on Learning Representations

    2023

  54. [62]

    Zhiqin Yang, Yonggang Zhang, Yu Zheng, Xinmei Tian, Hao Peng, Tongliang Liu, and Bo Han. 2024. FedFed: Feature distillation against data heterogeneity in federated learning. Advances in Neural Information Processing Systems 36 (2024)

    2024

  55. [63]

    Huatao Xu, Pengfei Zhou, Rui Tan, and Mo Li. 2023. Practically adopting human activity recognition. In Proceedings of the 29th Annual International Conference on Mobile Computing and Networking. 1–15

    2023

  56. [64]

    Michael Zhang, Karan Sapra, Sanja Fidler, Serena Yeung, and Jose M. Alvarez. 2021. Personalized Federated Learning with First Order Model Optimization. In International Conference on Learning Representations

    2021

  57. [65]

    Tianfei Zhou and Ender Konukoglu. 2023. FedF A: Federated Feature Augmentation. In The Eleventh International Conference on Learning Representations. 21

    2023

  58. [66]

    Rui Ye, Zhenyang Ni, Chenxin Xu, Jianyu Wang, Siheng Chen, and Yonina C. Eldar. 2023. FedFM: Anchor-Based Feature Matching for Data Heterogeneity in Federated Learning. IEEE Transactions on Signal Processing 71 (2023), 4224–4239. https://doi.org/10.1109/TSP.2023.3314277

    2023

  59. [2020]

    Advances in neural information processing systems 33 (2020), 9912–9924

    Unsupervised learning of visual features by contrasting cluster assignments. Advances in neural information processing systems 33 (2020), 9912–9924

    2020

  60. [2023]

    IEEE transactions on neural networks and learning systems (2023)

    Fedtp: Federated learning by transformer personalization. IEEE transactions on neural networks and learning systems (2023)

    2023

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.