Pith. sign in

REVIEW 3 major objections 5 minor 39 references

Federated Learning for Large-Scale Cloud Robotic Manipulation: Opportunities and Challenges

T0 review · 3 major / 5 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 This paper argues that federated learning, by training shared models across robot fleets while keeping raw data on each robot, is the key to scaling cloud robotic manipulation.

desk verdict A useful survey of FL for cloud robotic manipulation that overclaims novelty and skips the latency question; worth a round of revision. read the letter →

arxiv 2507.17903 v1 pith:ERASEKMY submitted 2025-07-23 cs.LG

Obaidullah Zaland , Chanh Nguyen , Florian T. Pokorny , Monowar Bhuyan This is my paper · ORCID
classification cs.LG
keywords federatedlearningcloudroboticsroboticmanipulationprivacy-preservingrobotfleetscloud-edgecontinuumcommunicationlatencyheterogeneity
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 tries to establish that federated learning (FL) is the key to addressing the challenges of large-scale cloud robotic manipulation: by training a shared model on each robot's local data and sending only model updates to a server, robot fleets could collaborate on manipulation skills without transferring raw, privacy-sensitive data. The authors believe this fusion could speed deployment of robotic fleets in domains such as healthcare and home assistance, where privacy is paramount. The paper surveys early federated robotic systems, catalogues the obstacles (latency, heterogeneity, security, limited resources, energy), and sets out research directions including clustered FL, interoperable FL, LLM integration, and responsible FL. As a position and survey paper, its intended contribution is a map of the opportunity space rather than a working system.

What carries the argument

The mechanism that carries the argument is the federated learning loop itself, anchored by the FedAvg aggregation rule. In the paper's description, a server initializes a global model, distributes it to participating robots, each robot trains for a fixed number of epochs on its local data, and the server computes a new global model as the data-size-weighted average of the returned local models, $\omega^t = \sum_{c \in C_m} \frac{n_c}{N}\omega^{t-1}_c$. This loop is what turns scattered, privacy-sensitive manipulation data into a shared policy without raw data transfer. The paper couples the loop with the cloud-edge continuum, where robots offload heavier computation to edge servers or the cloud, so the fleet itself becomes the distributed training population.

What would settle it

Run a physical pick-and-place experiment in which a fleet of manipulators performs a time-critical task under a live federated learning schedule over a realistic wireless network, and record each FL round-trip delay against the robot's control deadline for safe grasping; if the aggregation round trip exceeds the safe control loop bound or success rates collapse, the central claim fails.

Watch

Extended reading notes

Core claim

The paper's central claim is that federated learning is key and has the potential to address the challenges in cloud robotics and robotic manipulation. In its own telling, FL lets a fleet of robots learn a global manipulation model collaboratively: each robot trains locally on private data, the server aggregates the updated weights, and the loop repeats until convergence, so raw sensor data never leaves the robot. The authors further assert that this fusion is largely unstudied — no existing literature, they say, studies the combination of FL and cloud robotic manipulation — and that pursuing it would pave the way for trustful, private, robust, scalable, and efficient robotic systems. On that basis they present fundamental FL concepts, review early federated robotic efforts, and outline challenges and future research directions for realizing this vision at scale.

Load-bearing premise

The premise that holds everything together is that robots can tolerate the extra communication and aggregation delays of federated learning during manipulation without breaking real-time or safety constraints.

Editorial extensions

If this is right

  • Robot fleets could learn manipulation policies across hospitals or households without uploading patient records, home imagery, or other sensitive sensor streams.
  • Clustered FL could let fleets specialize: robots performing similar tasks form clusters that learn task-specific personalized models alongside a global model.
  • Interoperable FL could enable heterogeneous fleets with different sensors and architectures to share knowledge and generalize across environments.
  • Realizing the vision requires adaptive FL algorithms that cut communication load and tolerate stragglers, because aggregation delays directly threaten real-time manipulation.
  • Developing resource-efficient federated multimodal LLMs could give robots a shared reasoning layer for understanding and manipulating their surroundings.

Reading between the lines

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

  • Beyond the paper: the authors' assertion that no literature studies the fusion is contradicted by their own review, which describes several federated robotic systems; the field is young and fragmented rather than nonexistent.
  • A reader can infer that the first testable milestone is a small-scale physical deployment where a handful of manipulators train via FL over a realistic wireless link while performing time-bounded pick-and-place tasks, with success rate and round-trip latency measured.
  • The healthcare teleoperation example suggests a concrete pilot: hospital assistive robots trained collaboratively with FL while keeping patient data on-device, benchmarking privacy and task performance against centralized cloud training.
  • An implicit consequence is that FL's communication cost may shift the offloading decision: more computation may need to stay on the edge to keep aggregation rounds short enough for manipulation control loops.
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. This manuscript is a position/survey paper arguing that federated learning (FL) is key to enabling large-scale cloud robotic manipulation. It introduces FL fundamentals (centralized and decentralized FL, FedAvg, horizontal and vertical FL), describes the cloud-edge continuum and cloud robotics, reviews early FL-for-robotics works (SDRL, FLDDPG, PPAFL, federated imitation learning, LFRL), enumerates current challenges (communication/latency, heterogeneity, security/privacy, limited resources, energy), and sketches future directions (clustered FL, LLM integration, responsible FL, interoperability, security and trust). The paper contains no new algorithms, experiments, or derivations; its contribution is a framed research agenda and literature synthesis.

Significance. If the central premise were established, the paper would be a useful agenda-setting piece for privacy-preserving fleet learning in manipulation. Its challenge taxonomy is sensible, and the future directions (especially clustered FL and LLM-based manipulation policies) are timely. The paper does not provide machine-checked proofs, reproducible code, or falsifiable predictions, and its core novelty claim is internally inconsistent with the cited literature. Its value is therefore as a literature review and position statement rather than as a technical contribution.

major comments (3)
  1. [Section I] Section I states, 'As per our knowledge, no existing literature studies the aforementioned fusion,' but Section IV itself reviews [22], [25], and [26], and [26] is explicitly described as 'a starting point for the fusion of cloud robotics and FL.' This is an internal contradiction that overstates the paper's novelty. Please remove the claim or replace it with a precise statement of what is missing, e.g., no systematic treatment of manipulation-specific latency and safety constraints.
  2. [Sections I and V-A] The paper's central assertion that 'Federated learning is key and has the potential to address the challenges in cloud robotics and robotic manipulation' is load-bearing, yet no mechanism or evidence is given that FL's training-time rounds (model download, local training, upload, aggregation) can coexist with real-time manipulation control. Section V-A concedes that fleet size and data volume exacerbate communication and aggregation delays, and Sections V-D and V-E note latency and energy trade-offs, but the paper never provides a latency budget or argues that FL updates fit within manipulation deadlines (typically 10-100 Hz with safety constraints). At minimum, the paper should state that FL operates in the background and only periodically refreshes policies, and it should discuss how that timescale affects closed-loop manipulation; otherwise the thesis remains an unsupported assumption.
  3. [Section IV] The survey characterizations of prior work are sometimes stronger than the cited sources support. For example, [22] is described as a 'decentralized, federated reinforcement learning setup,' but the cited title is 'Swarm deep reinforcement learning for robotic manipulation' and does not mention federated learning; please verify the description against the original paper. Similarly, the FLDDPG discussion reports that the system 'overperforms centralized DDPG approaches' and 'significantly reduce communication costs extensively' without specifying the experimental setup or citing a particular figure or table. Since Section IV is the empirical basis for the claimed fusion, these characterizations need to be precise and traceable.
minor comments (5)
  1. [Section VI-A] The subsection on clustered FL is truncated mid-sentence: 'Clustered FL can further help distribute the communication load, as all the models would not be sent directly to the server. It can also' is immediately followed by Section VI-B, leaving the sentence incomplete. Please complete the thought or remove the dangling clause.
  2. [Section II, Eq. (1)] In Eq. (1), the symbols n_c, N, and C_m are not defined, making the aggregation formula difficult to parse. Additionally, FedAvg is commonly attributed to McMahan et al., 'Communication-Efficient Learning of Deep Networks from Decentralized Data,' not to the Konečný et al. reference [9] currently cited; please correct the citation.
  3. [Abstract] The final sentence of the abstract, 'where researchers adopt to design and verify FL models in either centralized or decentralized settings,' is ungrammatical and obscures the intended meaning; please rephrase.
  4. [Figure 1] Figure 1, captioned 'Centralized vs. federated cloud robotic setups,' is not referenced or explained in the body text; please refer to it where the centralized/decentralized distinction is introduced and clarify what each panel depicts.
  5. [References] Several references are incomplete or informal, including [15] (a Scribd URL with no venue) and [16] (a webpage with no author or access year); please standardize these entries to the journal's format.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity found: the paper contains no derivation chain, no fitted quantities, and its single self-citation is not load-bearing.

full rationale

This is a position/opportunities paper with no Eqs. (2)-(n) derivation chain, no fitted parameters, and no prediction that reduces to an input by construction. The only equation, FedAvg's weighted average (Eq. 1), is a standard definition quoted from the literature and is not used to derive any novel claim. The sole self-citation, ref. [3] (Zaland et al., one-shot FL with diffusion models), supports the general sentence 'federated learning (FL) [3] has evolved as a training methodology' (Section I); it is a background attribution, not a load-bearing premise for the paper's opportunity claims. The paper's assertation that 'no existing literature studies the aforementioned fusion' (Section I) is contradicted by the paper's own refs. [22], [25], and especially [26], which states it 'provided a starting point for the fusion of cloud robotics and FL' (Section IV); however, that is a novelty overclaim or factual inconsistency, not circular reasoning, because the cited works are independent external studies, not outputs of the present paper. Similarly, Section V-A's admission that robotic data volume and fleet size intensify communication and aggregation delays weakens the paper's central premise that FL is key for real-time cloud robotic manipulation, but a weak or unproven premise is a correctness risk, not a circular step: there is no fitted quantity renamed as a prediction and no result obtained from its own assumptions by definition. Under the stated rules, the honest finding is no significant circularity, score 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The paper's central claims rest on domain assumptions about FL's compatibility with real-time manipulation and about privacy preservation, not on any derived mathematics. No free parameters or invented entities appear; the assumptions are imported from the FL and robotics literature.

assumptions (3)
  • domain assumption Federated learning preserves privacy by keeping data on devices while still enabling useful model training (the FL paradigm's standard promise).
    Used throughout Sections III-B and IV to motivate FL as a privacy solution; the paper itself notes in Section V-C that model updates can leak information, so this premise is not unconditional.
  • domain assumption Robots can tolerate FL communication and aggregation delays during manipulation tasks without violating real-time or safety constraints.
    The opportunity claim in Sections I and VI presupposes feasible operation; Section V-A lists latency as a challenge, making this an open assumption rather than a demonstrated fact.
  • domain assumption The cited prior works accurately represent the state of FL for cloud robotics.
    A narrative survey relies on selected references; the paper does not provide a systematic search or inclusion criteria, so completeness cannot be checked.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Federated Learning for Large-Scale Cloud Robotic Manipulation: Opportunities and Challenges." pith.science (2026). https://pith.science/paper/ERASEKMY

@misc{pith2026250717903,
  author       = {Pith},
  title        = {Pith review of: Federated Learning for Large-Scale Cloud Robotic Manipulation: Opportunities and Challenges},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/ERASEKMY}},
  note         = {Machine review of arXiv:2507.17903}
}
read the original abstract

Federated Learning (FL) is an emerging distributed machine learning paradigm, where the collaborative training of a model involves dynamic participation of devices to achieve broad objectives. In contrast, classical machine learning (ML) typically requires data to be located on-premises for training, whereas FL leverages numerous user devices to train a shared global model without the need to share private data. Current robotic manipulation tasks are constrained by the individual capabilities and speed of robots due to limited low-latency computing resources. Consequently, the concept of cloud robotics has emerged, allowing robotic applications to harness the flexibility and reliability of computing resources, effectively alleviating their computational demands across the cloud-edge continuum. Undoubtedly, within this distributed computing context, as exemplified in cloud robotic manipulation scenarios, FL offers manifold advantages while also presenting several challenges and opportunities. In this paper, we present fundamental concepts of FL and their connection to cloud robotic manipulation. Additionally, we envision the opportunities and challenges associated with realizing efficient and reliable cloud robotic manipulation at scale through FL, where researchers adopt to design and verify FL models in either centralized or decentralized settings.

Figures

Figures reproduced from arXiv: 2507.17903 by the authors.

Figure 1
Figure 1. FIGURE 1 [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. FIGURE 2 [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 33 canonical work pages

  1. [22]

    Swarm deep reinforcement learning for robotic manipulation,

    X. Zhu, F. Zhang, and H. Li, “Swarm deep reinforcement learning for robotic manipulation,” Procedia Computer Science , vol. 198, pp. 472– 479, 2022

    work page 2022

  2. [25]

    Federated imitation learning: A novel framework for cloud robotic systems with heterogeneous sensor data,

    B. Liu, L. Wang, M. Liu, and C.-Z. Xu, “Federated imitation learning: A novel framework for cloud robotic systems with heterogeneous sensor data,” IEEE Robotics and Automation Letters , vol. 5, no. 2, pp. 3509– 3516, 2020

    work page 2020

  3. [26]

    Lifelong federated reinforcement learn- ing: a learning architecture for navigation in cloud robotic systems,

    B. Liu, L. Wang, and M. Liu, “Lifelong federated reinforcement learn- ing: a learning architecture for navigation in cloud robotic systems,” IEEE Robotics and Automation Letters , vol. 4, no. 4, pp. 4555–4562, 2019

    work page 2019

  4. [1]

    European Commission, “Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (General Data Protection Regulation) (Text with EEA relevance),”

    work page 2016

  5. [2]

    Federated learning application on depression treatment robots (dtbot),

    Y . Liu and R. Yang, “Federated learning application on depression treatment robots (dtbot),” in 2021 IEEE 13th International Conference on Computer Research and Development (ICCRD) . IEEE, 2021, pp. 121–124

    work page 2021

  6. [3]

    One-shot fed- erated learning with classifier-free diffusion models,

    O. Zaland, S. Jin, F. T. Pokorny, and M. Bhuyan, “One-shot fed- erated learning with classifier-free diffusion models,” arXiv preprint arXiv:2502.08488, 2025

    work page arXiv 2025

  7. [4]

    A survey of research on cloud robotics and automation,

    B. Kehoe, S. Patil, P. Abbeel, and K. Goldberg, “A survey of research on cloud robotics and automation,” IEEE Transactions on automation science and engineering , vol. 12, no. 2, pp. 398–409, 2015

    work page 2015

  8. [5]

    Toward robotic manipulation,

    M. T. Mason, “Toward robotic manipulation,” Annual Review of Con- trol, Robotics, and Autonomous Systems , vol. 1, pp. 1–28, 2018

    work page 2018

Show all 39 references
  1. [6]

    Robofleet: Open source communication and management for fleets of autonomous robots,

    K. S. Sikand, L. Zartman, S. Rabiee, and J. Biswas, “Robofleet: Open source communication and management for fleets of autonomous robots,” in 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE, 2021, pp. 406–412

    2021

  2. [7]

    Petuum: A new platform for distributed machine learning on big data,

    X. et al., “Petuum: A new platform for distributed machine learning on big data,” in Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining , 2015, pp. 1335– 1344

    2015

  3. [8]

    From distributed machine learning to federated learning: In the view of data privacy and security,

    S. Shen, T. Zhu, D. Wu, W. Wang, and W. Zhou, “From distributed machine learning to federated learning: In the view of data privacy and security,” Concurrency and Computation: Practice and Experience , vol. 34, no. 16, p. e6002, 2022

    2022

  4. [9]

    Federated learning: Strategies for improving communication efficiency,

    J. Kone ˇcn`y, H. B. McMahan, F. X. Yu, P. Richt ´arik, A. T. Suresh, and D. Bacon, “Federated learning: Strategies for improving communication efficiency,” arXiv preprint arXiv:1610.05492 , 2016

    2016 arXiv

  5. [10]

    Federated optimization in heterogeneous networks,

    T. Li, A. K. Sahu, M. Zaheer, M. Sanjabi, A. Talwalkar, and V . Smith, “Federated optimization in heterogeneous networks,” Proceedings of Machine learning and systems , vol. 2, pp. 429–450, 2020

    2020

  6. [11]

    Pravfed: Practical heterogeneous vertical federated learning via representation learning,

    S. Wang, K. Gai, J. Yu, Z. Zhang, and L. Zhu, “Pravfed: Practical heterogeneous vertical federated learning via representation learning,” IEEE Transactions on Information F orensics and Security , 2025

    2025

  7. [12]

    Edge computing: Vision and challenges,

    W. Shi, J. Cao, Q. Zhang, Y . Li, and L. Xu, “Edge computing: Vision and challenges,” IEEE internet of things journal , vol. 3, no. 5, pp. 637– 646, 2016

    2016

  8. [13]

    The seminal role of edge-native applications,

    M. Satyanarayanan, G. Klas, M. Silva, and S. Mangiante, “The seminal role of edge-native applications,” in 2019 IEEE International Confer- ence on Edge Computing (EDGE) . IEEE, 2019, pp. 33–40

    2019

  9. [14]

    Distributed intelligence on the edge-to-cloud continuum: A systematic literature review,

    D. Rosendo, A. Costan, P. Valduriez, and G. Antoniu, “Distributed intelligence on the edge-to-cloud continuum: A systematic literature review,” Journal of Parallel and Distributed Computing , vol. 166, pp. 71–94, 2022

    2022

  10. [15]

    Cloud-enabled humanoid robotics,

    J. Kuffner, “Cloud-enabled humanoid robotics,” http://www. scribd. com/doc/47896204/James-Kuffner-Humanoids2010, 2010, accessed: 2023-09-30

    2010

  11. [16]

    A. A. RoboMaker, https://aws.amazon.com/robomaker/, 2018, accessed: 2023-09-30

    2018

  12. [17]

    Rapyuta: A cloud robotics platform,

    G. Mohanarajah, D. Hunziker, R. D’Andrea, and M. Waibel, “Rapyuta: A cloud robotics platform,” IEEE Transactions on Automation Science and Engineering , vol. 12, no. 2, pp. 481–493, 2014

    2014

  13. [18]

    Fogros 2: An adaptive and extensible platform for cloud and fog robotics using ros 2,

    J. Ichnowski, K. Chen, K. Dharmarajan, S. Adebola, M. Danielczuk, V . Mayoral-Vilches, H. Zhan, D. Xu, R. Ghassemi, J. Kubiatowicz, et al. , “Fogros 2: An adaptive and extensible platform for cloud and fog robotics using ros 2,” in Proceedings IEEE International Conference on ...

    2023

  14. [19]

    Kuberos: A unified platform for auto- mated and scalable deployment of ros2-based multi-robot applications,

    Y . Zhang, C. Wurll, and B. Hein, “Kuberos: A unified platform for auto- mated and scalable deployment of ros2-based multi-robot applications,” in 2023 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2023, pp. 9097–9103

    2023

  15. [20]

    X-mas: Extremely large-scale multi-modal sensor dataset for outdoor surveillance in real environments,

    D. Noh, C. Sung, T. Uhm, W. Lee, H. Lim, J. Choi, K. Lee, D. Hong, D. Um, I. Chung, et al. , “X-mas: Extremely large-scale multi-modal sensor dataset for outdoor surveillance in real environments,” IEEE Robotics and Automation Letters , vol. 8, no. 2, pp. 1093–1100, 2023

    2023

  16. [21]

    Deep reinforcement learning: An overview,

    Y . Li, “Deep reinforcement learning: An overview,” arXiv preprint arXiv:1701.07274, 2017

    2017 arXiv

  17. [23]

    Federated reinforcement learning for collective navigation of robotic swarms,

    S. Na, T. Rou ˇcek, J. Ulrich, J. Pikman, T. s Krajn ´ık, B. Lennox, and F. Arvin, “Federated reinforcement learning for collective navigation of robotic swarms,” IEEE Transactions on Cognitive and Developmental Systems, 2023

    2023

  18. [24]

    Decentralized p2p federated learning for privacy-preserving and resilient mobile robotic systems,

    X. Zhou, W. Liang, K. I.-K. Wang, Z. Yan, L. T. Yang, W. Wei, J. Ma, and Q. Jin, “Decentralized p2p federated learning for privacy-preserving and resilient mobile robotic systems,” IEEE Wireless Communications , vol. 30, no. 2, pp. 82–89, 2023

    2023

  19. [27]

    Federated learning with gan-based data synthesis for non-iid clients,

    Z. Li, J. Shao, Y . Mao, J. H. Wang, and J. Zhang, “Federated learning with gan-based data synthesis for non-iid clients,” in International Workshop on Trustworthy Federated Learning . Springer, 2022, pp. 17–32

    2022

  20. [28]

    Cooperative heterogeneous multi-robot systems: A survey,

    Y . Rizk, M. Awad, and E. W. Tunstel, “Cooperative heterogeneous multi-robot systems: A survey,” ACM Computing Surveys (CSUR) , vol. 52, no. 2, pp. 1–31, 2019

    2019

  21. [29]

    Security for industrial communication systems,

    D. Dzung, M. Naedele, T. P. V on Hoff, and M. Crevatin, “Security for industrial communication systems,” Proceedings of the IEEE , vol. 93, no. 6, pp. 1152–1177, 2005

    2005

  22. [30]

    Healthcare robotics,

    L. D. Riek, “Healthcare robotics,” Communications of the ACM, vol. 60, no. 11, pp. 68–78, 2017

    2017

  23. [31]

    Ldp-fl: Practical private aggregation in federated learning with local differential privacy,

    L. Sun, J. Qian, and X. Chen, “Ldp-fl: Practical private aggregation in federated learning with local differential privacy,” arXiv preprint arXiv:2007.15789, 2020

    2007 arXiv

  24. [32]

    Hacking robots before skynet,

    C. Cerrudo and L. Apa, “Hacking robots before skynet,” IOActive Website, pp. 1–17, 2017

    2017

  25. [33]

    Resource allocation of federated learning for the metaverse with mobile augmented reality,

    X. Zhou, C. Liu, and J. Zhao, “Resource allocation of federated learning for the metaverse with mobile augmented reality,” IEEE Transactions on Wireless Communications , 2023

    2023

  26. [34]

    A hybrid model for energy-efficient green internet of things enabled intelligent transportation systems using federated learning,

    S. Kaleem, A. Sohail, M. Babar, A. Ahmad, and M. U. Tariq, “A hybrid model for energy-efficient green internet of things enabled intelligent transportation systems using federated learning,” Internet of Things , vol. 25, p. 101038, 2024

    2024

  27. [35]

    Clustered federated multi-task learning with non-iid data,

    Y . Xiao, J. Shu, X. Jia, and H. Huang, “Clustered federated multi-task learning with non-iid data,” in 2021 IEEE 27th International Conference on Parallel and Distributed Systems (ICPADS). IEEE, 2021, pp. 50–57

    2021

  28. [36]

    A survey of resource-efficient llm and multimodal foundation models,

    M. Xu, W. Yin, D. Cai, R. Yi, D. Xu, Q. Wang, B. Wu, Y . Zhao, C. Yang, S. Wang, et al. , “A survey of resource-efficient llm and multimodal foundation models,” arXiv preprint arXiv:2401.08092 , 2024

    2024 arXiv

  29. [37]

    Ifedavg: Interpretable data-interoperability for federated learning,

    D. Roschewitz, M.-A. Hartley, L. Corinzia, and M. Jaggi, “Ifedavg: Interpretable data-interoperability for federated learning,” arXiv preprint arXiv:2107.06580, 2021

    2021 arXiv

  30. [38]

    Establishing trust in cloud computing,

    K. M. Khan and Q. Malluhi, “Establishing trust in cloud computing,” IT professional, vol. 12, no. 5, pp. 20–27, 2010

    2010

  31. [2016]

    Available: https://eur-lex.europa.eu/eli/reg/2016/679/oj

    [Online]. Available: https://eur-lex.europa.eu/eli/reg/2016/679/oj

    2016

Pith tools

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