arxiv: 2605.00011 · v1 · submitted 2026-03-11 · 💻 cs.LG · cs.AI· cs.DC

Recognition: no theorem link

FedACT: Concurrent Federated Intelligence across Heterogeneous Data Sources

Md Sirajul Islam , Isabelle G Chapman , N I Md Ashafuddula , Xu Yuan , Li Chen , Nian-Feng Tzeng , Klara Nahrstedt

Authors on Pith no claims yet

Pith reviewed 2026-05-15 12:28 UTC · model grok-4.3

classification 💻 cs.LG cs.AIcs.DC

keywords federated learningdevice schedulingheterogeneous resourcesconcurrent jobsjob completion timeresource allocationparticipation fairness

0 comments

The pith

FedACT schedules heterogeneous devices across multiple concurrent federated learning jobs using alignment scores to minimize average job completion time.

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

The paper introduces FedACT as a scheduling method for running several federated learning tasks simultaneously on shared devices with varying resources. It evaluates how well each device's available compute, memory, and bandwidth match a job's demands through an alignment score, then assigns devices accordingly while enforcing balanced participation across jobs. This dual focus on compatibility and fairness aims to reduce the time until all jobs finish their training rounds. Experiments on benchmark datasets show large gains in speed and final model accuracy over prior approaches that treat jobs in isolation. If the method generalizes, shared device pools could support many learning tasks at once without one job starving the others of suitable hardware.

Core claim

FedACT formulates an optimal scheduling plan that prioritizes devices with higher alignment scores between their available resources and each job's resource demands, while incorporating participation fairness constraints to balance device contributions across concurrent FL jobs, resulting in lower average job completion time and higher global model accuracy than baselines.

What carries the argument

The alignment scoring mechanism that evaluates compatibility between a device's available resources and a job's resource demands.

Load-bearing premise

The alignment scoring mechanism accurately captures compatibility between dynamic device resources and job demands, and enforcing participation fairness does not materially increase overall job completion time.

What would settle it

A controlled test with real heterogeneous devices running several simultaneous FL jobs that measures whether the reported reductions in average job completion time and accuracy gains persist against the same baselines.

Figures

Figures reproduced from arXiv: 2605.00011 by Isabelle G Chapman, Klara Nahrstedt, Li Chen, Md Sirajul Islam, Nian-Feng Tzeng, N I Md Ashafuddula, Xu Yuan.

**Figure 2.** Figure 2: An overview of the training procedure within the [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Figure 3: Test accuracy versus elapsed wall-clock time for different jobs in Group A with the IID distribution. [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗

**Figure 4.** Figure 4: Test accuracy versus elapsed wall-clock time for different jobs in Group B with the IID distribution. [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗

**Figure 5.** Figure 5: Test accuracy versus elapsed wall-clock time for different jobs in Group A with the Non-IID distribution. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗

**Figure 6.** Figure 6: Test accuracy versus elapsed wall-clock time for different jobs in Group B with the Non-IID distribution. [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗

**Figure 7.** Figure 7: Elapsed wall-clock training time required for each job of Group A to achieve the target convergence accuracy under [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗

**Figure 8.** Figure 8: Elapsed wall-clock training time required for each job of Group B to achieve the target convergence accuracy under [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗

read the original abstract

Federated Learning (FL) enables collaborative intelligence across decentralized data source devices in a privacy-preserving way. While substantial research attention has been drawn to optimizing the learning process for an individual task, real-world applications increasingly require multiple machine learning tasks simultaneously training their models across a shared pool of devices. Naively applying single-FL optimization techniques in multi-FL systems results in suboptimal system performance, particularly due to device heterogeneity and resource inefficiency. To address such a critical open challenge, we introduce {\em FedACT}, a novel resource heterogeneity-aware device scheduling approach designed to efficiently schedule heterogeneous devices across multiple concurrent FL jobs, with the goal of minimizing their average job completion time (JCT). {\em FedACT} dynamically assigns devices to FL jobs based on an alignment scoring mechanism that evaluates the compatibility between available resources of devices and resource demands of jobs. Additionally, it incorporates participation fairness to ensure balanced contributions from devices across jobs, further enhancing the accuracy levels of learned global models. An optimal scheduling plan is formulated in {\em FedACT} by prioritizing devices with higher alignment scores, while ensuring fair participation across jobs. To evaluate the effectiveness of the proposed scheduling algorithm, we carried out comprehensive experiments using diverse FL jobs and benchmark datasets. Experimental results demonstrate that {\em FedACT} reduces the average JCT by up to 8.3\(\times\) and improves model accuracy by up to 44.5\%, compared to the state-of-the-art baselines.

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.

Desk Editor's Note private letter to a colleague

FedACT gives a workable greedy scheduler for multi-job FL on heterogeneous devices but the 8.3× gains rest on an unproven heuristic that may be trace-specific.

read the letter

FedACT introduces a scheduler that scores how well each device's resources match the demands of several concurrent federated learning jobs, then assigns devices greedily while adding a fairness rule to balance participation. The stated goal is lower average job completion time across the jobs. That framing is new for the multi-job setting, where most prior work still optimizes one model at a time. The paper does a reasonable job of spelling out why single-task techniques fall short when devices are shared and heterogeneous, and the experiments use several benchmark datasets rather than one narrow case. Those elements make the contribution concrete for people who actually run multiple models on edge hardware. The main soft spot is the scheduling rule itself. The paper writes down an optimal assignment but solves it by ordering devices on the alignment score and picking in that order, with no approximation guarantee or competitive analysis. When job demands are correlated or device availability changes quickly, the greedy choice can miss better global assignments, so the reported 8.3× reduction in completion time and 44.5 % accuracy lift could shrink on other traces. The fairness term is added to protect model quality, yet its separate cost in total time is not isolated. The experimental claims are large enough that the paper needs to show the baselines were re-implemented fairly and that the gains survive different seeds or device distributions. If those checks are present and clean in the full text, the numbers become more believable. This work is aimed at systems researchers who deploy federated learning on real heterogeneous pools rather than theorists. A reader who needs a ready-to-code multi-job scheduler will find usable pieces. It deserves a serious referee because the problem is timely and the method is testable, though the analysis would need tightening before acceptance.

Referee Report

2 major / 2 minor

Summary. The paper introduces FedACT, a resource heterogeneity-aware device scheduling approach for concurrent federated learning jobs across heterogeneous devices. It uses an alignment scoring mechanism to evaluate compatibility between device resources and job demands, incorporates a participation fairness term, and formulates an optimal plan by prioritizing high-score devices. Experiments on diverse FL jobs and benchmark datasets claim reductions in average job completion time by up to 8.3× and model accuracy improvements by up to 44.5% versus state-of-the-art baselines.

Significance. If the results hold under rigorous verification, FedACT addresses a practical challenge in multi-task FL by improving efficiency and model quality in resource-heterogeneous settings. The combination of dynamic scoring and fairness is a useful systems contribution, though the absence of approximation guarantees for the greedy scheduler limits broader impact.

major comments (2)

[Abstract] Abstract: The abstract asserts large empirical gains (up to 8.3× JCT reduction and 44.5% accuracy improvement) but supplies no description of the alignment scoring formula, experimental setup, baseline implementations, statistical tests, or potential post-hoc choices. This omission is load-bearing because it prevents any assessment of whether the data actually support the central claims.
[Scheduling algorithm] Scheduling algorithm: The method formulates an optimal plan but solves it via greedy selection ordered by alignment score plus fairness term, with no proof or approximation bound showing the result is within a constant factor of optimality. This is load-bearing for the JCT minimization claim, as greedy choice properties can fail under correlated job demands or high device churn, turning reported speedups into potentially setup-specific outcomes.

minor comments (2)

[Abstract] Abstract: Raw LaTeX commands such as {em FedACT} appear in the text; these should be rendered as proper formatting in the final manuscript.
[Method] Throughout: Provide explicit mathematical definitions or pseudocode for the alignment score and the fairness constraint to improve reproducibility and clarity.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments on our manuscript. We address each major point below with honest responses and indicate where revisions will be incorporated.

read point-by-point responses

Referee: [Abstract] Abstract: The abstract asserts large empirical gains (up to 8.3× JCT reduction and 44.5% accuracy improvement) but supplies no description of the alignment scoring formula, experimental setup, baseline implementations, statistical tests, or potential post-hoc choices. This omission is load-bearing because it prevents any assessment of whether the data actually support the central claims.

Authors: We acknowledge that the abstract is high-level and omits key details on the alignment scoring formula, experimental setup, baselines, and statistical procedures. These elements are fully described in Sections 3.2 (alignment scoring in Equation 3), 4.1 (setup, datasets, and baseline implementations), and 4.2 (results with averages over 5 independent runs and standard deviations). To address the concern directly, we will revise the abstract to include a concise sentence on the alignment scoring mechanism and fairness term while noting that full experimental details appear in the body. This is a partial revision due to abstract length limits. revision: partial
Referee: [Scheduling algorithm] Scheduling algorithm: The method formulates an optimal plan but solves it via greedy selection ordered by alignment score plus fairness term, with no proof or approximation bound showing the result is within a constant factor of optimality. This is load-bearing for the JCT minimization claim, as greedy choice properties can fail under correlated job demands or high device churn, turning reported speedups into potentially setup-specific outcomes.

Authors: The referee is correct that we formulate the scheduling objective as an optimization problem but solve it via greedy selection on alignment scores plus the fairness term, without providing an approximation guarantee. The underlying assignment problem is NP-hard under general resource constraints and dynamic arrivals, which precludes a simple constant-factor proof. We will revise Section 3.3 to explicitly label the scheduler as a greedy heuristic, discuss its potential limitations under correlated demands or high churn, and add new experiments simulating those conditions to demonstrate that the reported gains remain consistent. This strengthens the presentation without overstating optimality. revision: partial

Circularity Check

0 steps flagged

No significant circularity detected in derivation chain

full rationale

The paper introduces FedACT as a new algorithmic scheduling method based on an alignment scoring mechanism for device-job compatibility plus a fairness term. No equations, fitted parameters, or derivation steps appear in the provided text that reduce by construction to prior inputs, self-citations, or renamed empirical patterns. The central claims rest on experimental comparisons rather than a closed mathematical loop; the greedy prioritization of an 'optimal plan' is presented as a constructive heuristic without self-referential definitions or load-bearing self-citations that would force the reported JCT gains. This is a standard case of an independent algorithmic proposal evaluated empirically.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Review performed on abstract only; the central claim rests on an unspecified alignment scoring function and an unstated fairness constraint whose precise definitions and parameterizations are not provided.

axioms (2)

domain assumption Device resource availability can be meaningfully scored for compatibility with job resource demands
Invoked by the alignment scoring mechanism described in the abstract.
domain assumption Enforcing balanced device participation across jobs improves or at least does not harm global model accuracy
Stated as part of the design goal in the abstract.

pith-pipeline@v0.9.0 · 5582 in / 1285 out tokens · 36824 ms · 2026-05-15T12:28:20.837817+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

50 extracted references · 50 canonical work pages · 4 internal anchors

[1]

EU. 2018. European Union’s General Data Protection Regulation (GDPR). European Union. Accessed 2024-04. [Online]. Available: https://eugdpr.org/

work page 2018
[2]

Communication-efficient learning of deep networks from decentralized data,

B. McMahan, E. Moore, D. Ramage, S. Hampson, and B. A. y Arcas, “Communication-efficient learning of deep networks from decentralized data,” inArtificial intelligence and statistics. PMLR, 2017, pp. 1273– 1282

work page 2017
[4]

Falcon: Addressing stragglers in heterogeneous parameter server via multiple parallelism,

Q. Zhou, S. Guo, H. Lu, L. Li, M. Guo, Y . Sun, and K. Wang, “Falcon: Addressing stragglers in heterogeneous parameter server via multiple parallelism,”IEEE Transactions on Computers, vol. 70, no. 1, pp. 139– 155, 2020

work page 2020
[5]

Clusterfl: a similarity-aware federated learning system for human activity recogni- tion,

X. Ouyang, Z. Xie, J. Zhou, J. Huang, and G. Xing, “Clusterfl: a similarity-aware federated learning system for human activity recogni- tion,” inProceedings of the 19th Annual International Conference on Mobile Systems, Applications, and Services, 2021, pp. 54–66

work page 2021
[6]

Federated Learning for Mobile Keyboard Prediction

A. Hard, K. Rao, R. Mathews, S. Ramaswamy, F. Beaufays, S. Augen- stein, H. Eichner, C. Kiddon, and D. Ramage, “Federated learning for mobile keyboard prediction,”arXiv preprint arXiv:1811.03604, 2018

work page internal anchor Pith review Pith/arXiv arXiv 2018
[7]

Oort: Efficient federated learning via guided participant selection,

F. Lai, X. Zhu, H. V . Madhyastha, and M. Chowdhury, “Oort: Efficient federated learning via guided participant selection,” inOSDI, 2021, pp. 19–35

work page 2021
[8]

Pyramidfl: A fine-grained client selection framework for efficient federated learning,

C. Li, X. Zeng, M. Zhang, and Z. Cao, “Pyramidfl: A fine-grained client selection framework for efficient federated learning,” inProceedings of the 28th annual international conference on mobile computing and networking, 2022, pp. 158–171

work page 2022
[9]

Fedfairˆ3: Unlocking threefold fairness in federated learning,

S. Javaherian, S. Panta, S. Williams, M. S. Islam, and L. Chen, “Fedfairˆ3: Unlocking threefold fairness in federated learning,”in Pro- ceedings of IEEE International Conference on Communications (ICC), pp. 1–7, 2024

work page 2024
[10]

Fedclust: Tackling data heterogeneity in federated learning through weight-driven client clustering,

M. S. Islam, S. Javaherian, F. Xu, X. Yuan, L. Chen, and N.-F. Tzeng, “Fedclust: Tackling data heterogeneity in federated learning through weight-driven client clustering,” inProceedings of the 53rd International Conference on Parallel Processing, 2024, pp. 474–483

work page 2024
[11]

Communication-efficient federated learning via knowledge distillation,

C. Wu, F. Wu, L. Lyu, Y . Huang, and X. Xie, “Communication-efficient federated learning via knowledge distillation,”Nature communications, vol. 13, no. 1, p. 2032, 2022

work page 2032
[12]

Fedboost: A communication- efficient algorithm for federated learning,

J. Hamer, M. Mohri, and A. T. Suresh, “Fedboost: A communication- efficient algorithm for federated learning,” inInternational Conference on Machine Learning. PMLR, 2020, pp. 3973–3983

work page 2020
[13]

Ditto: Fair and robust federated learning through personalization,

T. Li, S. Hu, A. Beirami, and V . Smith, “Ditto: Fair and robust federated learning through personalization,” inInternational Conference on Machine Learning. PMLR, 2021, pp. 6357–6368

work page 2021
[14]

Pgfed: Personalize each client’s global objective for federated learning,

J. Luo, M. Mendieta, C. Chen, and S. Wu, “Pgfed: Personalize each client’s global objective for federated learning,”International Confer- ence on Computer Vision, 2023

work page 2023
[15]

Fedala: Adaptive local aggregation for personalized federated learning,

J. Zhang, Y . Hua, H. Wang, T. Song, Z. Xue, R. Ma, and H. Guan, “Fedala: Adaptive local aggregation for personalized federated learning,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 37, no. 9, 2023, pp. 11 237–11 244

work page 2023
[16]

Asynchronous federated optimization,

C. Xie, S. Koyejo, and I. Gupta, “Asynchronous federated optimization,” arXiv preprint arXiv:1903.03934, 2019

work page arXiv 1903
[17]

Federated learning with buffered asynchronous aggregation,

J. Nguyen, K. Malik, H. Zhan, A. Yousefpour, M. Rabbat, M. Malek, and D. Huba, “Federated learning with buffered asynchronous aggregation,” inInternational Conference on Artificial Intelligence and Statistics. PMLR, 2022, pp. 3581–3607

work page 2022
[18]

Fedcore: Straggler-free federated learning with distributed coresets,

H. Guo, H. Gu, X. Wang, B. Chen, E. K. Lee, T. Eilam, D. Chen, and K. Nahrstedt, “Fedcore: Straggler-free federated learning with distributed coresets,” inICC 2024-IEEE International Conference on Communications. IEEE, 2024, pp. 280–286

work page 2024
[19]

Seafl: Enhancing efficiency in semi-asynchronous federated learning through adaptive aggregation and selective training,

M. S. Islam, S. Panta, F. Xu, X. Yuan, L. Chen, and N.-F. Tzeng, “Seafl: Enhancing efficiency in semi-asynchronous federated learning through adaptive aggregation and selective training,” in2025 IEEE International Parallel and Distributed Processing Symposium (IPDPS), 2025, pp. 509– 519

work page 2025
[20]

Efficient device scheduling with multi-job federated learning,

C. Zhou, J. Liu, J. Jia, J. Zhou, Y . Zhou, H. Dai, and D. Dou, “Efficient device scheduling with multi-job federated learning,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 36, no. 9, 2022, pp. 9971–9979

work page 2022
[21]

Fedast: federated asynchronous simultaneous training,

B. Askin, P. Sharma, C. Joe-Wong, and G. Joshi, “Fedast: federated asynchronous simultaneous training,”arXiv preprint arXiv:2406.00302, 2024

work page arXiv 2024
[22]

Learning multiple layers of features from tiny images,

A. Krizhevsky, G. Hintonet al., “Learning multiple layers of features from tiny images,” 2009

work page 2009
[23]

Gradient-based learning applied to document recognition,

Y . LeCun, L. Bottou, Y . Bengio, and P. Haffner, “Gradient-based learning applied to document recognition,”Proceedings of the IEEE, vol. 86, no. 11, pp. 2278–2324, 1998

work page 1998
[24]

Emnist: Extending mnist to handwritten letters,

G. Cohen, S. Afshar, J. Tapson, and A. Van Schaik, “Emnist: Extending mnist to handwritten letters,” in2017 international joint conference on neural networks (IJCNN). IEEE, 2017, pp. 2921–2926

work page 2017
[25]

Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms

H. Xiao, K. Rasul, and R. V ollgraf, “Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms,”arXiv preprint arXiv:1708.07747, 2017

work page internal anchor Pith review Pith/arXiv arXiv 2017
[26]

Federated learning: Challenges, methods, and future directions,

T. Li, A. K. Sahu, A. Talwalkar, and V . Smith, “Federated learning: Challenges, methods, and future directions,”IEEE signal processing magazine, vol. 37, no. 3, pp. 50–60, 2020

work page 2020
[27]

Scaffold: Stochastic controlled averaging for federated learn- ing,

S. P. Karimireddy, S. Kale, M. Mohri, S. Reddi, S. Stich, and A. T. Suresh, “Scaffold: Stochastic controlled averaging for federated learn- ing,” inInternational Conference on Machine Learning. PMLR, 2020, pp. 5132–5143

work page 2020
[28]

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

work page 2020
[29]

Public-key cryptosystems based on composite degree residu- osity classes,

P. Paillier, “Public-key cryptosystems based on composite degree residu- osity classes,” inInternational conference on the theory and applications of cryptographic techniques. Springer, 1999, pp. 223–238

work page 1999
[30]

Practical Secure Aggregation for Federated Learning on User-Held Data

K. Bonawitz, V . Ivanov, B. Kreuter, A. Marcedone, H. B. McMa- han, S. Patel, D. Ramage, A. Segal, and K. Seth, “Practical secure aggregation for federated learning on user-held data,”arXiv preprint arXiv:1611.04482, 2016

work page internal anchor Pith review Pith/arXiv arXiv 2016
[31]

Differential privacy: A survey of results,

C. Dwork, “Differential privacy: A survey of results,” inInternational conference on theory and applications of models of computation. Springer, 2008, pp. 1–19

work page 2008
[32]

Joint device scheduling and resource allocation for latency constrained wireless federated learn- ing,

W. Shi, S. Zhou, Z. Niu, M. Jiang, and L. Geng, “Joint device scheduling and resource allocation for latency constrained wireless federated learn- ing,”IEEE Transactions on Wireless Communications, vol. 20, no. 1, pp. 453–467, 2020

work page 2020
[33]

Speeding up distributed machine learning using codes,

K. Lee, M. Lam, R. Pedarsani, D. Papailiopoulos, and K. Ramchandran, “Speeding up distributed machine learning using codes,”IEEE Trans- actions on Information Theory, vol. 64, no. 3, pp. 1514–1529, 2017

work page 2017
[34]

Load distribution fairness in p2p data management systems,

T. Pitoura and P. Triantafillou, “Load distribution fairness in p2p data management systems,” in2007 IEEE 23rd International Conference on Data Engineering. IEEE, 2006, pp. 396–405

work page 2006
[35]

“fairness analysis

A. Finkelstein, M. Harman, S. A. Mansouri, J. Ren, and Y . Zhang, ““fairness analysis” in requirements assignments,” in2008 16th IEEE International Requirements Engineering Conference. IEEE, 2008, pp. 115–124

work page 2008
[36]

A multi- agent q-learning-based framework for achieving fairness in http adaptive streaming,

S. Petrangeli, M. Claeys, S. Latr ´e, J. Famaey, and F. De Turck, “A multi- agent q-learning-based framework for achieving fairness in http adaptive streaming,” in2014 IEEE Network Operations and Management Sym- posium (NOMS). IEEE, 2014, pp. 1–9

work page 2014
[37]

A survey on bias and fairness in machine learning,

N. Mehrabi, F. Morstatter, N. Saxena, K. Lerman, and A. Galstyan, “A survey on bias and fairness in machine learning,”ACM computing surveys (CSUR), vol. 54, no. 6, pp. 1–35, 2021

work page 2021
[38]

Very Deep Convolutional Networks for Large-Scale Image Recognition

K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,”arXiv preprint arXiv:1409.1556, 2014

work page internal anchor Pith review Pith/arXiv arXiv 2014
[39]

Imagenet classification with deep convolutional neural networks,

A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,”Advances in neural informa- tion processing systems, vol. 25, 2012

work page 2012
[40]

Deep residual learning for image recognition,

K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” inProceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770–778

work page 2016
[41]

Cellular traffic load prediction with lstm and gaussian process regression,

W. Wang, C. Zhou, H. He, W. Wu, W. Zhuang, and X. Shen, “Cellular traffic load prediction with lstm and gaussian process regression,” inICC 2020-2020 IEEE international conference on communications (ICC). IEEE, 2020, pp. 1–6

work page 2020
[42]

Scheduling algorithms for efficient execution of stream workflow applications in multicloud environments,

M. Barika, S. Garg, A. Chan, and R. N. Calheiros, “Scheduling algorithms for efficient execution of stream workflow applications in multicloud environments,”IEEE transactions on services computing, vol. 15, no. 2, pp. 860–875, 2019

work page 2019
[43]

Towards federated learning at scale: System design,

K. Bonawitz, H. Eichner, W. Grieskamp, D. Huba, A. Ingerman, V . Ivanov, C. Kiddon, J. Kone ˇcn`y, S. Mazzocchi, B. McMahanet al., 11 Accepted at the 40th IEEE International Parallel & Distributed Processing Symposium (IPDPS), 2026, New Orleans, USA “Towards federated learning at scale: System design,”Proceedings of machine learning and systems, vol. 1, p...

work page 2026
[44]

Fedvision: An online visual object detection platform powered by federated learning,

Y . Liu, A. Huang, Y . Luo, H. Huang, Y . Liu, Y . Chen, L. Feng, T. Chen, H. Yu, and Q. Yang, “Fedvision: An online visual object detection platform powered by federated learning,” inProceedings of the AAAI Conference on Artificial Intelligence, vol. 34, no. 08, 2020, pp. 13 172– 13 179

work page 2020
[45]

To talk or to work: Flexible communication compression for energy efficient federated learning over heterogeneous mobile edge devices,

L. Li, D. Shi, R. Hou, H. Li, M. Pan, and Z. Han, “To talk or to work: Flexible communication compression for energy efficient federated learning over heterogeneous mobile edge devices,” inIEEE INFOCOM 2021-IEEE Conference on Computer Communications. IEEE, 2021, pp. 1–10

work page 2021
[46]

Enabling privacy-preserving incentives for mobile crowd sensing systems,

H. Jin, L. Su, B. Ding, K. Nahrstedt, and N. Borisov, “Enabling privacy-preserving incentives for mobile crowd sensing systems,” in 2016 IEEE 36th International Conference on Distributed Computing Systems (ICDCS). IEEE, 2016, pp. 344–353

work page 2016
[47]

Multi-job intelligent scheduling with cross-device federated learning,

J. Liu, J. Jia, B. Ma, C. Zhou, J. Zhou, Y . Zhou, H. Dai, and D. Dou, “Multi-job intelligent scheduling with cross-device federated learning,” IEEE Transactions on Parallel and Distributed Systems, vol. 34, no. 2, pp. 535–551, 2022

work page 2022
[48]

Joint participant selection and learning scheduling for multi-model federated edge learning,

X. Wei, J. Liu, and Y . Wang, “Joint participant selection and learning scheduling for multi-model federated edge learning,” in2022 IEEE 19th International Conference on Mobile Ad Hoc and Smart Systems (MASS). IEEE, 2022, pp. 537–545

work page 2022
[49]

Multi-model federated learning,

N. Bhuyan and S. Moharir, “Multi-model federated learning,” in2022 14th International Conference on COMmunication Systems & NETworkS (COMSNETS). IEEE, 2022, pp. 779–783

work page 2022
[50]

Fair training of multiple federated learning models on resource constrained network devices,

M. Siew, S. Arunasalam, Y . Ruan, Z. Zhu, L. Su, S. Ioannidis, E. Yeh, and C. Joe-Wong, “Fair training of multiple federated learning models on resource constrained network devices,” inProceedings of the 22nd International Conference on Information Processing in Sensor Networks, 2023, pp. 330–331

work page 2023
[51]

Venn: Re- source management across federated learning jobs,

J. Liu, F. Lai, D. Ding, Y . Zhang, and M. Chowdhury, “Venn: Re- source management across federated learning jobs,”arXiv preprint arXiv:2312.08298, 2023. 12

work page arXiv 2023