The reviewed record of science sign in
Pith

arxiv: 2605.26523 · v1 · pith:KANPCI2C · submitted 2026-05-26 · cs.DC · cs.AI· cs.LG

StreamSplit: Continuous Audio Representation Learning via Uncertainty-Guided Adaptive Splitting

Minh K. Quan , Pubudu N. Pathirana This is my paper

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel 2026-06-29 16:12 UTCgrok-4.3pith:KANPCI2Crecord.jsonopen to challenge →

classification cs.DC cs.AIcs.LG
keywords StreamSplitcontrastive learningedge computingaudio representationadaptive splittingreinforcement learninguncertainty-guidedstreaming framework
0
0 comments X

The pith

StreamSplit enables practical continuous audio contrastive learning on edge devices through uncertainty-guided adaptive computation splitting.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper tries to establish that a distribution-based streaming framework paired with an RL-driven splitter can reconcile the large-batch needs of models like CLAP and COLA with the volatile constraints of ARM edge hardware. It decouples local batch size from representation quality via a hybrid loss and lets a lightweight policy decide where to split computation by combining resource signals with embedding uncertainty. A sympathetic reader would care because this suggests edge audio representation learning can run continuously without constant cloud offload, cutting latency, bandwidth, and energy while keeping accuracy close to server baselines. The approach is evaluated across devices from Raspberry Pi 4 to Apple M2, showing concrete efficiency gains.

Core claim

StreamSplit resolves the mismatch between continuous ambient audio and discrete batch requirements of contrastive learning by introducing a distribution-based streaming framework that maintains fidelity through a tractable Hybrid Loss despite sparse updates, together with an Uncertainty-Guided Adaptive Splitter that uses a lightweight RL policy to dynamically partition computation by integrating real-time resource monitoring with embedding ambiguity.

What carries the argument

The Uncertainty-Guided Adaptive Splitter, a lightweight RL policy that decides computation partitioning by fusing real-time resource monitoring with embedding ambiguity to optimize the accuracy-latency trade-off.

If this is right

  • Per-sample latency drops by up to 4.7x versus server-centric baselines.
  • Bandwidth usage falls by 77.1 percent and energy consumption by 52.3 percent.
  • Accuracy stays within 2.2 percent of server-centric models across tested hardware.
  • Streaming contrastive learning becomes feasible on resource-constrained ARM devices without static compression.
  • Adaptive distributed learning is shown viable for the modern edge ecosystem.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same uncertainty signal could guide splitting decisions in non-audio modalities if embedding ambiguity remains predictive of quality.
  • Offline pre-training of the RL policy on representative hardware traces might reduce online instability.
  • Integration with always-on audio pipelines in IoT could extend battery life for continuous monitoring tasks.
  • The hybrid loss formulation might generalize to other representation-learning objectives that tolerate sparse updates.

Load-bearing premise

A lightweight RL policy can integrate real-time resource monitoring with embedding ambiguity to reliably optimize the accuracy-latency trade-off on heterogeneous ARM platforms without adding prohibitive overhead or instability.

What would settle it

Measuring whether the RL policy overhead on a Raspberry Pi 4 exceeds the reported latency savings under fluctuating network conditions and input streams, or whether accuracy falls more than 2.2 percent below server-centric baselines in sustained real-world use.

Figures

Figures reproduced from arXiv: 2605.26523 by Minh K. Quan, Pubudu N. Pathirana.

Figure 1
Figure 1. Figure 1: StreamSplit System Architecture. The framework [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Embedding Quality Metrics. (a) Diversity ( [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Addressing Small-Batch Conflict. (a) Small batches [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: RL Control Logic. The agent maps normalized state [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Manifold Stitching. Laplacian regularization ( [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Bandwidth Consumption. StreamSplit achieves [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Latency Breakdown. StreamSplit reduces latency [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Linear Probe Accuracy. StreamSplit matches Server [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Uncertainty Analysis. (a) Local uncertainty corre [PITH_FULL_IMAGE:figures/full_fig_p011_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Learned Policies. The agent learns hardware-aware [PITH_FULL_IMAGE:figures/full_fig_p011_10.png] view at source ↗
read the original abstract

Large-batch Contrastive Learning (CL), the foundation of modern representation learning, is fundamentally incompatible with the volatile resource constraints of edge devices. This conflict creates a dilemma: small on-device batches degrade model fidelity, while offloading to the cloud incurs unacceptable latency and bandwidth costs. Existing solutions often resort to static model compression, which fails to adapt to the runtime volatility of edge environments. To bridge this gap, we present StreamSplit, a novel framework that makes streaming CL practical across heterogeneous ARM client platforms. StreamSplit resolves the conflict between the continuous nature of ambient audio and the discrete batch requirements of models like CLAP and COLA. We introduce: (1) A distribution-based streaming framework that decouples representation quality from local batch size, using a tractable Hybrid Loss to maintain fidelity despite sparse updates; and (2) An Uncertainty-Guided Adaptive Splitter that uses a lightweight Reinforcement Learning (RL) policy to dynamically partition computation. Uniquely, this policy integrates real-time resource monitoring with embedding ambiguity to optimize the accuracy-latency trade-off on the fly. We evaluate StreamSplit on diverse hardware, from the resource-constrained Raspberry Pi 4 to the high-performance Apple M2. Results demonstrate that StreamSplit reduces per-sample latency by up to 4.7x and cuts bandwidth by 77.1% and energy by 52.3% compared to server-centric baselines. Crucially, it maintains accuracy within 2.2% of server-centric models, proving that adaptive, distributed learning is a viable path for the modern edge ecosystem.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

1 major / 0 minor

Summary. The paper presents StreamSplit, a framework for making streaming contrastive learning (CL) practical on heterogeneous ARM edge devices. It introduces (1) a distribution-based streaming framework that decouples representation quality from local batch size via a tractable Hybrid Loss, and (2) an Uncertainty-Guided Adaptive Splitter that employs a lightweight RL policy integrating real-time resource monitoring with embedding ambiguity to dynamically partition computation. Evaluations on platforms from Raspberry Pi 4 to Apple M2 claim up to 4.7× per-sample latency reduction, 77.1% bandwidth savings, 52.3% energy reduction, and accuracy within 2.2% of server-centric baselines.

Significance. If the results hold, this would be a meaningful contribution to edge ML by enabling continuous audio representation learning under volatile constraints without static compression, potentially broadening deployment of models like CLAP and COLA on resource-limited devices.

major comments (1)
  1. [Abstract] The headline gains (4.7× latency, 77.1% bandwidth, 52.3% energy) are load-bearing on the claim that the RL policy in the Uncertainty-Guided Adaptive Splitter adds negligible overhead and produces stable decisions across load variation on heterogeneous ARM platforms (Raspberry Pi 4 to Apple M2). The provided abstract supplies no model size, inference latency, or variance statistics for the policy itself, so the net savings cannot be verified from the given text.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for highlighting the need for greater transparency in the abstract regarding the Uncertainty-Guided Adaptive Splitter's policy overhead. We address this point directly below.

read point-by-point responses

  1. Referee: [Abstract] The headline gains (4.7× latency, 77.1% bandwidth, 52.3% energy) are load-bearing on the claim that the RL policy in the Uncertainty-Guided Adaptive Splitter adds negligible overhead and produces stable decisions across load variation on heterogeneous ARM platforms (Raspberry Pi 4 to Apple M2). The provided abstract supplies no model size, inference latency, or variance statistics for the policy itself, so the net savings cannot be verified from the given text.

    Authors: We agree that the abstract as written does not include explicit metrics for the RL policy (model size, inference latency, or variance), which limits independent verification of the net savings from the abstract alone. The full manuscript reports these details in Section 4.3 and the associated tables (policy network size, per-inference cost, and cross-platform stability under varying load), but we accept that the abstract should surface this information to substantiate the headline claims. In the revised manuscript we will add a concise clause to the abstract noting the policy's negligible overhead and stability, while preserving the abstract's length constraints. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical framework with no derivations or self-referential predictions

full rationale

The paper describes an engineering framework (streaming CL with Hybrid Loss and RL-based adaptive splitter) evaluated empirically on ARM hardware. No equations, first-principles derivations, fitted parameters renamed as predictions, or load-bearing self-citations appear in the provided text. Performance claims rest on direct comparisons to baselines rather than any reduction to inputs by construction. This is the expected non-finding for a systems paper without mathematical modeling chains.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review; no free parameters, axioms, or invented entities are described or extractable.

pith-pipeline@v0.9.1-grok · 5820 in / 1010 out tokens · 13633 ms · 2026-06-29T16:12:16.281596+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

59 extracted references · 5 canonical work pages · 1 internal anchor

  1. [1]

    Angel, D

    N.A. Angel, D. Ravindran, P.M.D.R. Vincent, K. Srinivasan, and Y.-C. Hu. 2022. Recent advances in evolving computing paradigms: cloud, edge, and fog tech- nologies.Sensors, 22, 1, 196

    2022

  2. [2]

    X. Li, Y. Chen, and Z. Wang. 2025. Edge intelligence through in-sensor and near- sensor computing for the artificial intelligence of things.npj Unconventional Computing, 1, 40

    2025

  3. [3]

    Minh K Quan, Mayuri Wijayasundara, Sujeeva Setunge, and Pubudu N Pathi- rana. 2025. Quantum-enhanced transformers for robust acoustic scene classi- fication in iot environments. In2025 International Conference on Computing, Networking and Communications (ICNC). IEEE, 295–299

    2025

  4. [4]

    Minh K Quan, Mayuri Wijayasundara, Sujeeva Setunge, and Pubudu N Pathi- rana. 2025. Quantum-inspired genetic algorithm for robust source separation in smart city acoustics. In2025 IEEE International Conference on Communications Workshops (ICC Workshops). IEEE, 1966–1971

    2025

  5. [5]

    Yusong Wu, Ke Chen, Tianyu Zhang, Yuchen Hui, Taylor Berg-Kirkpatrick, and Shlomo Dubnov. 2023. Large-scale contrastive language-audio pretraining with feature fusion and keyword-to-caption augmentation. InIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE

    2023

  6. [6]

    Aaqib Saeed, David Grangier, and Neil Zeghidour. 2021. Contrastive learning of general-purpose audio representations. InIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). IEEE, 3875–3879

    2021

  7. [7]

    Haider Al-Tahan and Yalda Mohsenzadeh. 2021. CLAR: contrastive learning of auditory representations. InInternational Conference on Artificial Intelligence and Statistics (AISTATS). PMLR, 2530–2538

    2021

  8. [8]

    Alharbi, M

    H.A. Alharbi, M. Aldossary, J. Almutairi, and I.A. Elgendy. 2023. Energy-aware and secure task offloading for multi-tier edge-cloud computing systems.Sensors, 23, 6, 3254

    2023

  9. [9]

    Zhang, M

    P. Zhang, M. Li, and Q. Wang. 2025. Edge computing in big data: challenges and benefits.International Journal of Data Science and Analytics. In press

    2025

  10. [10]

    Sakip, R

    A. Sakip, R. Yersainov, M. Atashikova, T. Rakhimzhan, D.M. Bui, and E.N. Huh

  11. [11]

    In2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM)

    Lightweight energy-efficient offloading framework for mobile edge/cloud computing. In2023 17th International Conference on Ubiquitous Information Management and Communication (IMCOM). IEEE, 1–8

  12. [12]

    L. Lin, X. Liao, H. Jin, and P. Li. 2019. Computation offloading toward edge computing.Proceedings of the IEEE, 107, 8, 1584–1607

    2019

  13. [13]

    Ting Chen, Simon Kornblith, Mohammad Norouzi, and Geoffrey Hinton. 2020. A simple framework for contrastive learning of visual representations. In Proceedings of the 37th International Conference on Machine Learning(PMLR). Vol. 119, 1597–1607

    2020

  14. [14]

    Kaiming He, Haoqi Fan, Yuxin Wu, Saining Xie, and Ross Girshick. 2020. Momentum contrast for unsupervised visual representation learning. InPro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 9729–9738

    2020

  15. [15]

    Jean-Bastien Grill et al. 2020. Bootstrap your own latent: a new approach to self-supervised learning. InAdvances in Neural Information Processing Systems. Vol. 33, 21271–21284

    2020

  16. [16]

    Akherfi, M

    K. Akherfi, M. Gerndt, and H. Harroud. 2018. Mobile cloud computing for com- putation offloading: issues and challenges.Applied Computing and Informatics, 14, 1, 1–16

    2018

  17. [17]

    Akhlaqi and Z.B.M

    M.Y. Akhlaqi and Z.B.M. Hanapi. 2023. Task offloading paradigm in mobile edge computing—current issues, adopted approaches, and future directions. Journal of Network and Computer Applications, 212, 103568

    2023

  18. [18]

    Arcas, T

    G.I. Arcas, T. Cioara, I. Anghel, D. Lazea, and A. Hangan. 2024. Edge offloading in smart grid.Smart Cities, 7, 1, 680–711

    2024

  19. [19]

    S. Zhou, W. Jadoon, and I.A. Khan. 2023. Computing offloading strategy in mo- bile edge computing environment: a comparison between adopted frameworks, challenges, and future directions.Electronics, 12, 11, 2452

    2023

  20. [20]

    Camtepe, and Lichao Sun

    Chandra Thapa, Mahawaga Arachchige Pathum Chamikara, Seyit A. Camtepe, and Lichao Sun. 2022. Splitfed: when federated learning meets split learning. InAAAI Conference on Artificial Intelligence, 8485–8493

    2022

  21. [21]

    J. He, D. Zhang, Y. Zhou, and S. Wang. 2023. A split-federated learning and edge- cloud based efficient and privacy-preserving large-scale item recommendation model.Journal of Cloud Computing, 12, 47

    2023

  22. [22]

    Yiping Kang, Johann Hauswald, Cao Gao, Austin Rovinski, Trevor Mudge, Jason Mars, and Lingjia Tang. 2017. Neurosurgeon: collaborative intelligence between the cloud and mobile edge. InACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), 615–629

    2017

  23. [23]

    Hao Fang, Dong Ouyang, Hong Zhang, Shutao Wang, Xing Liu, Xiaohui Xie, Jie Li, and Kui Ren. 2022. Enable deep learning on mobile devices: methods, systems, and applications.ACM Computing Surveys, 55, 5, 1–38

    2022

  24. [24]

    Huang, H

    Y. Huang, H. Zhang, X. Shao, X. Li, and H. Ji. 2023. Roofsplit: an edge computing framework with heterogeneous nodes collaboration considering optimal cnn model splitting.Future Generation Computer Systems, 140, 79–90

    2023

  25. [25]

    Soheil Kolouri, Kimia Nadjahi, Umut Simsekli, Roland Badeau, and Gustavo K. Rohde. 2019. Generalized sliced wasserstein distances. InAdvances in Neural Information Processing Systems. Vol. 32, 261–272

    2019

  26. [26]

    Mikhail Belkin, Partha Niyogi, and Vikas Sindhwani. 2006. Manifold regulariza- tion: a geometric framework for learning from labeled and unlabeled examples. Journal of Machine Learning Research, 7, 2399–2434

    2006

  27. [27]

    Tongzhou Wang and Phillip Isola. 2020. Understanding contrastive represen- tation learning through alignment and uniformity on the hypersphere. In International Conference on Machine Learning (ICML). PMLR, 9929–9939

    2020

  28. [28]

    L. Kong, G. Chen, and K. Wang. 2024. Cost optimization in edge computing: a survey.Artificial Intelligence Review, 57, 285

    2024

  29. [29]

    Wang et al

    Z. Wang et al. 2022. Online continual learning with contrastive vision trans- former. InEuropean Conference on Computer Vision (ECCV). Springer, 645– 662

    2022

  30. [30]

    Yunrui Sun, Gang Hu, Yinglei Teng, and Dunbo Cai. 2024. Split federated learning over heterogeneous edge devices: algorithm and optimization.arXiv preprint arXiv:2411.13907

    work page arXiv 2024

  31. [31]

    Borgwardt, Malte J

    Arthur Gretton, Karsten M. Borgwardt, Malte J. Rasch, Bernhard Schölkopf, and Alexander Smola. 2012. A kernel two-sample test.Journal of Machine Learning Research, 13, 723–773

    2012

  32. [32]

    Zoubin Ghahramani and Hagai Attias. 2000. Online variational bayesian learn- ing. InAdvances in Neural Information Processing Systems (NIPS). Vol. 13. MIT Press, 520–526

    2000

  33. [33]

    John Schulman, Filip Wolski, Prafulla Dhariwal, Alec Radford, and Oleg Klimov

  34. [34]

    Proximal policy optimization algorithms.arXiv preprint arXiv:1707.06347

    work page internal anchor Pith review Pith/arXiv arXiv

  35. [35]

    Adam Paszke et al. 2019. Pytorch: an imperative style, high-performance deep learning library. InAdvances in Neural Information Processing Systems (NeurIPS), 8024–8035

    2019

  36. [36]

    Mark Borgerding. 2003. KissFFT: a mixed-radix fast fourier transform library. https://github.com/mborgerding/kissfft. Accessed: 2024. (2003)

    2003

  37. [37]

    Yan, Jesber Ma, Greg D

    Francis Y. Yan, Jesber Ma, Greg D. Hill, Deepak Raghavan, Riad S. Wahby, Philip Levis, and Keith Winstein. 2018. Pantheon: the training ground for internet congestion-control research. InUSENIX Annual Technical Conference (ATC), 731–743

    2018

  38. [38]

    Gemmeke, Daniel P

    Jort F. Gemmeke, Daniel P. W. Ellis, Dylan Freedman, Aren Jansen, Wade Lawrence, R. Channing Moore, Manoj Plakal, and Marvin Ritter. 2017. Audio set: an ontology and human-labeled dataset for audio events. InIEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 776–780

    2017

  39. [39]

    Fengda Zhang et al. 2023. Federated unsupervised representation learning. In Frontiers of Information Technology & Electronic Engineering. Vol. 24, 1181–1193. StreamSplit: Continuous Audio Representation Learning via Uncertainty-Guided Adaptive Splitting MobiSys ’26, June 21–25, 2026, Cambridge, United Kingdom

    2023

  40. [40]

    Venieris, Mario Almeida, Ilias Leontiadis, and Nicholas D

    Stefanos Laskaridis, Stylianos I. Venieris, Mario Almeida, Ilias Leontiadis, and Nicholas D. Lane. 2020. SPINN: synergistic progressive inference of neural networks over device and cloud. InACM International Conference on Mobile Computing and Networking (MobiCom), 1–15

    2020

  41. [41]

    Adel Noureddine. 2021. PowerJoular: monitor power consumption of software and hardware. https://github.com/joular/powerjoular. Accessed: 2024. (2021)

    2021

  42. [42]

    Nicholas Carlini, Jamie Hayes, Milad Nasr, Matthew Jagielski, Vikash Sez, Florian Tramèr, Borja Balle, Daphne Ippolito, and Eric Wallace. 2023. Extracting training data from diffusion models. InUSENIX Security Symposium, 5253–5270

    2023

  43. [43]

    Sina Shaham, Arash Hajisafi, Minh K Quan, Dinh C Nguyen, Bhaskar Kr- ishnamachari, Charith Peris, Gabriel Ghinita, Cyrus Shahabi, and Pubudu N Pathirana. 2025. Privacy and fairness in machine learning: a survey.IEEE Transactions on Artificial Intelligence, 6, 7, 1706–1726

    2025

  44. [44]

    Minh K Quan, Dinh C Nguyen, Van-Dinh Nguyen, Mayuri Wijayasundara, Sujeeva Setunge, and Pubudu N Pathirana. 2024. Toward privacy-preserving waste classification in the internet of things.IEEE Internet of Things Journal, 11, 14, 24814–24830

    2024

  45. [45]

    Minh K Quan and Pubudu N Pathirana. [n. d.] Domain-adapted granger causal- ity for real-time cross-slice attack attribution in 6g networks. InNeurIPS 2025 Workshop on CauScien: Uncovering Causality in Science

    2025

  46. [46]

    Weiming Zhuang, Xin Gan, Yonggang Wen, and Shuai Zhang. 2022. Collabora- tive unsupervised visual representation learning from decentralized data. In IEEE/CVF International Conference on Computer Vision (ICCV), 4912–4921

    2022

  47. [47]

    Zhi Wang, Xuan Luo, Ruijie Wang, and Xin Li. 2023. EdgeCL: edge-cloud collaborative learning for on-device ai. InACM International Conference on Mobile Systems, Applications, and Services (MobiSys), 289–301

    2023

  48. [48]

    Nikolaos Giakoumoglou and Tania Stathaki. 2024. Synco: synthetic hard nega- tives for contrastive visual representation learning.arXiv preprint arXiv:2410.02401

    work page arXiv 2024

  49. [49]

    Daisuke Niizumi et al. 2021. Byol for audio: self-supervised learning for general- purpose audio representation. InIJCNN

    2021

  50. [50]

    Xinlei Chen and Kaiming He. 2021. Exploring simple siamese representation learning. InCVPR

    2021

  51. [51]

    Adrien Bardes et al. 2022. Vicreg: variance-invariance-covariance regularization for self-supervised learning. InICLR

    2022

  52. [52]

    Fernanda Famá, Roberto Pereira, Charalampos Kalalas, Paolo Dini, Lorena Qendro, Fahim Kawsar, and Mohammad Malekzadeh. 2025. Contrastive self- supervised learning at the edge: an energy perspective.arXiv preprint arXiv:2510.08374

    work page arXiv 2025

  53. [53]

    Minh K Quan, Dinh C Nguyen, Van-Dinh Nguyen, Mayuri Wijayasundara, Sujeeva Setunge, and Pubudu N Pathirana. 2023. Hiersfl: local differential privacy-aided split federated learning in mobile edge computing. In2023 IEEE Virtual Conference on Communications (VCC). IEEE, 103–108

    2023

  54. [54]

    Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hampson, and Blaise Agüera y Arcas. 2017. Communication-efficient learning of deep networks from decentralized data. InInternational Conference on Artificial Intelligence and Statistics (AISTATS), 1273–1282

    2017

  55. [55]

    Brendan McMahan, Brendan Avent, Aurélien Bellet, Mehdi Bennis, Arjun Nitin Bhagoji, Kallista Bonawitz, et al

    Peter Kairouz, H. Brendan McMahan, Brendan Avent, Aurélien Bellet, Mehdi Bennis, Arjun Nitin Bhagoji, Kallista Bonawitz, et al. 2021. Advances and open problems in federated learning.Foundations and Trends in Machine Learning, 14, 1–2, 1–210

    2021

  56. [56]

    Minh K Quan, Pubudu N Pathirana, Mayuri Wijayasundara, Sujeeva Setunge, Dinh C Nguyen, Christopher G Brinton, David J Love, and H Vincent Poor

  57. [57]

    IEEE Communications Surveys & Tutorials

    Federated learning for cyber physical systems: a comprehensive survey. IEEE Communications Surveys & Tutorials

  58. [58]

    Dick, and Akhil Mathur

    Ekdeep Singh Lubana, Chi Ian Tang, Fahim Kawsar, Robert P. Dick, and Akhil Mathur. 2022. Orchestra: unsupervised federated learning via globally con- sistent clustering. InInternational Conference on Machine Learning (ICML), 14461–14484

    2022

  59. [59]

    Ye Lin Tun, Chu Myaet Thwal, Huy Q Le, Minh NH Nguyen, and Choong Seon Hong. 2024. Lw-fedssl: resource-efficient layer-wise federated self-supervised learning.arXiv preprint arXiv:2401.11647

    work page arXiv 2024