Challenges, Applications and Design Aspects of Federated Learning: A Survey
Open accessAbstract
Federated Learning (FL) is a new technology that has been a hot research topic. It enables training an algorithm across multiple decentralized edge devices or servers holding local data samples, without exchanging them. There are many application domains where large amounts of properly labeled and complete data are not available in a centralized location, for example, doctors’ diagnosis from medical image analysis. There are also growing concerns over data and user privacy as Artificial Intelligence is becoming ubiquitous in new application domains. As such, very recently, a lot of research has been conducted in several areas within the nascent field of FL. A variety of surveys on different subtopics exist in current literature, focusing on specific challenges, design aspects and application domains. In this paper, we review existing contemporary works in the related areas in order to understand the challenges and topics that are emphasized by each type of FL surveys. Furthermore, we categorize FL research in terms of challenges, design factors and applications, conducting a holistic review of each and outlining promising research directions.
AI snapshot
Written from the paper’s full text.
TL;DR
This survey reviews contemporary federated learning (FL) survey papers and organizes FL research into design aspects, core challenges, and application areas. It provides a holistic review of each topic and outlines open issues and future research directions.
Key findings
- The reviewed surveys do not cover all subtopics; in particular, less than half thoroughly reviewed FL architectures and personalization techniques, motivating this paper’s broader classification.
- Design aspects span data partitioning (horizontal, vertical, federated transfer learning), architectures (client-server and peer-to-peer), aggregation algorithms (e.g., FedAvg, Scaffold, Adaptive Federated Optimization, FedBoost, FedProx, FedMA, Secure Aggregation), and personalization techniques (e.g., user context, clustering, transfer learning, multi-task/meta-learning, knowledge distillation, base+personalization layers, mixture of global/local models).
- Core challenges are communication cost, systems heterogeneity, statistical heterogeneity, and privacy/security; the paper summarizes mitigation strategies such as local updating, compression, decentralized training, asynchronous communication, client participation, fault tolerance, SMC, DP, and HE, and notes FL models are more vulnerable to model poisoning than data poisoning.
- Application areas covered include NLP, healthcare, computer vision, and transportation; open issues include tradeoffs among accuracy, privacy, communication cost, and personalization, and the lack of clear metrics for personalization.
Methods
- Survey/review design: the authors investigated and analyzed contemporary FL survey papers, then classified FL research into design aspects, core challenges, and application areas, with comparison tables for surveys, aggregation algorithms, personalization techniques, communication-cost strategies, systems-heterogeneity strategies, and threat models.
- No primary experiments or new datasets are reported; the work synthesizes existing literature and outlines open issues and future directions.
Limitations
- Stated in paper: tradeoffs among accuracy, privacy, communication cost, and level of personalization must be carefully considered and are often use-case dependent; no clear metrics have yet been formulated to measure personalization performance; implementation challenges for different application areas have not been thoroughly studied, with current focus primarily on training FL models.
- Evident from text: as a survey, it provides no empirical validation or quantitative benchmarking of its own, and its coverage is limited to the reviewed contemporary surveys and selected works.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
0 of 68 cited sources checked · 0 supported · 0 flagged
- Findingcitation
Scaffold solves the problem of client drifting using variance reduction technique in its local update.
“Scaffold (Stochastic Controlled Averaging Federated Learning) [58] solves the problem of client drifting using variance reduction technique in its local update.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [58]
Not checked in the free run.
- Findingcitation
FedAvg has convergence issues in some settings due to factors like client drifting and lack of adaptive learning rate.
“Although FedAvg has achieved great success, it has some convergence issues in some settings due to the factors like client drifting [58] and lack of adaptive learning rate [128].”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [58]
Not checked in the free run.
- cites [128] source
Not checked in the free run.
- Findingown data
Less than half of the reviewed survey papers thoroughly reviewed FL architectures and personalization techniques.
“In particular, less than half of the surveys thoroughly reviewed FL architectures and personalization techniques.”
Quote in the paperSubstantial claim - Findingnone given
Peer-to-peer FL architecture requires more computation for message encryption and decryption.
“However, it requires more computation for message encryption and decryption.”
Quote in the paperNot a substantial claim - Findingown data
The reviewed survey papers do not cover all the subtopics as highlighted in Table II.
“The reviewed survey papers do not cover all the subtopics as highlighted in Table II.”
Quote in the paperSubstantial claim
- Findingcitation
Adaptive Federated Optimization does not completely remove the effect of client heterogeneity.
“However, it does not completely remove the effect of client heterogeneity.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [128] source
Not checked in the free run.
- Findingcitation
FedMA can handle client heterogeneity.
“FedMA can also handle client heterogeneity.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [110]
Not checked in the free run.
- Findingcitation
FedBoost is a communication-efficient algorithm for FL based on ensemble learning.
“FedBoost [129] is a communication-efficient algorithm for FL based on ensemble learning technique.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [129] source
Not checked in the free run.
- Findingcitation
Within a few rounds of training, FedMA performs better than FedProx and FedAvg.
“Within a few rounds of training, it performs better than FedProx and FedAvg.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [110]
Not checked in the free run.
- Findingcitation
Both meta-learning and multi-task learning algorithms have been proposed in the federated setting to achieve greater personalization.
“Both meta-learning [99] and multi-task learning [100], [131], [132] algorithms have been proposed in the federated setting to achieve greater personalization.”
Quote in the paperNot a substantial claim0 of 3 sources checked- cites [99] source
Not checked in the free run.
- cites [100]
Not checked in the free run.
- cites [131]
Not checked in the free run.
- Findingreasoning
Adaptive Federated Optimization is compatible with cross-device FL.
“Moreover, it is compatible with cross-device FL.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [128] source
Not checked in the free run.
- Findingcitation
Secure Aggregation is developed based on the principle of Secure Multiparty Computation.
“Secure Aggregation [19] algorithm is developed based on the principle of Secure Multiparty Computation (SMC) algorithm.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [19]
Not checked in the free run.
- Findingcitation
Hierarchical communication patterns reduce the dependency on central server.
“Hierarchical communication patterns [142] is another approach that reduces the dependency on central server.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [142]
Not checked in the free run.
- Findingnone given
Local-updating schemes tackle communication costs by performing additional work on the client that generates and consumes the machine learning model.
“Local-updating schemes tackle communication costs by performing additional work on the client that generates and consumes the machine learning model.”
Quote in the paperNot a substantial claim - Findingcitation
FedProx enables partial work to be tolerated.
“First, it enables partial work to be tolerated.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [4]
Not checked in the free run.
- Findingcitation
Adaptive Federated Optimization does not increase client storage or communication costs.
“Although it incorporates adaptive learning rates in the server optimization, it does not increase client storage or communication costs.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [128] source
Not checked in the free run.
- Findingcitation
Transfer learning can be used in a federated setting for model personalization.
“Transfer learning can also be used in a federated setting for model personalization [130].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [130] source
Not checked in the free run.
- Findingcitation
Sketched and structured updates are compression schemes that enable reduction of the model update size communicated to the FL server from participating clients during each round.
“Sketched and structures updates are among the compression schemes that enable the reduction of the model update size communicated to the FL server from the participating clients during each round [26], [139].”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [26]
Not checked in the free run.
- cites [139]
Not checked in the free run.
- Findingcitation
The authors in [27] extend the work of [26] to reduce the cost of communication from the server to participant, employing approaches such as federated dropout and lossy compression.
“The authors in [27] further extend the work of [26] to reduce the cost of communication from the server to participant, employing approaches such as federated dropout and lossy compression.”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [27]
Not checked in the free run.
- cites [26]
Not checked in the free run.
- Findingcitation
Distributed local-updating primal methods have been applied with success for both convex and non-convex objectives.
“For both convex and non-convex objectives, distributed local-updating primal methods have also been applied with success [137].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [137] source
Not checked in the free run.
- Findingcitation
DLG obtained the private training data at the inference phase.
“Deep Leakage from Gradients (DLG) [152] obtained the private training data at the inference phase.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [152] source
Not checked in the free run.
- Findingcitation
Authors in [29] have come up with CMFL algorithm, which reduces the cost of communication by only uploading updates of the local model that are relevant.
“Authors in [29] have come up with the Communication-Mitigated Federated Learning (CMFL) algorithm, which reduces the cost of communication by only uploading updates of the local model that are relevant.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [29] source
Not checked in the free run.
- Findingnone given
Asynchronous communication suffers from bounded-delay assumptions made to control the measure of staleness.
“However, asynchronous communication also suffers from boundeddelay assumptions made to control the measure of staleness.”
Quote in the paperNot a substantial claim - Findingcitation
The FEDPER approach outperforms a pure FedAvg approach using the Flickr-AES dataset.
“The authors empirically show that the FEDPER approach outperforms a pure FedAvg approach using the Flickr-AES dataset [134] considering the personalization layers are capable of representing the personal predilection of an FL user.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [134] source
Not checked in the free run.
- Findingcitation
When exchanging gradient, private information of participant may be exposed to the adversary.
“When exchanging gradient, private information of participant may be exposed to the adversary [62], [150]-[152].”
Quote in the paperSubstantial claim0 of 3 sources checked- cites [62]
Not checked in the free run.
- cites [150]
Not checked in the free run.
- cites [151]
Not checked in the free run.
- Findingcitation
Global convergence is still guaranteed in CMFL.
“Global convergence is still guaranteed.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [29] source
Not checked in the free run.
- Findingcitation
Eichner et al. developed a pluralistic solution to alleviate a form of data heterogeneity where devices exhibited different characteristics during the day versus at night.
“Eichner et al. [95] developed a pluralistic solution to alleviate a form of data heterogeneity where devices exhibited different characteristics during the day versus at night.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [95]
Not checked in the free run.
- Findingcitation
iDLG exposes the label of training inputs.
“Another algorithm, iDLG also expose the label of training inputs [154].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [154]
Not checked in the free run.
- Findingnone given
No clear metrics have yet been formulated to measure the performance of personalization techniques.
“Nevertheless, no clear metrics have yet been formulated to measure the performance of personalization techniques.”
Quote in the paperNot a substantial claim - Findingnone given
FL models are more vulnerable for model poisoning attack than data poisoning attack.
“FL models are more vulnerable for model poisoning attack than data poisoning attack.”
Quote in the paperNot a substantial claim - Findingcitation
FL has been implemented successfully in several real-life applications.
“Although FL faces some limitations and severe challenges, it has been implemented successfully in several real-life applications:”
Quote in the paperNot a substantial claim - Findingnone given
The data generated and collected by devices in the network are usually non-identically distributed.
“The data that is generated and collected by the devices in the network are usually nonidentically distributed.”
Quote in the paperNot a substantial claim - Findingreasoning
For some clients, the global shared model does not perform as well as local models that are trained locally.
“For some clients, the global shared model does not perform as well as local models that are trained locally.”
Quote in the paperSubstantial claim - Findingcitation
Client participation schemes involve actively selecting participating devices based on systems resources like FedCS and quality of data at each round.
“Client participation schemes involve actively selecting participating devices based on systems resources like FedCS [33] and quality of data [47] at each round.”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [33]
Not checked in the free run.
- cites [47]
Not checked in the free run.
- Findingreasoning
Due to differences in connectivity, memory, CPU, battery power, participants in a federated network often have varying capacity in terms of communication, computation and storage.
“Due to having differences in connectivity of network, memory, CPU, battery power level, etc., participants in a federated network often have varying capacity in terms of communication, computation and storage.”
Quote in the paperSubstantial claim - Findingcitation
It is observed from [169] that a gap exists between the accuracy of personalized and global models.
“Moreover, it is observed from [169] that a gap exists between the accuracy of personalized and global models, making a case for personalization techniques to be one of the important research areas in FL.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [169]
Not checked in the free run.
- Findingcitation
Hitaj et al. apply GAN attack which allows the adversarial party in the training process of FL to fabricate inferring class representative.
“Hitaj et al. [155] apply GAN attack which allows the adversarial party in the training process of FL to fabricate inferring class representative.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [155] source
Not checked in the free run.
- Findingreasoning
When there is device variability, synchronous approaches are more susceptible to stragglers.
“When there is device variability, synchronous approaches are more susceptible to stragglers.”
Quote in the paperNot a substantial claim - Findingcitation
Various algorithms have been proposed to address systems heterogeneity.
“Various algorithms [33], [35] have been proposed to address systems heterogeneity.”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [33]
Not checked in the free run.
- cites [35] source
Not checked in the free run.
- Findingcitation
Sample-specific privacy guarantee technique by Li et al. trade off privacy for a greater accuracy.
“Sample-specific privacy guarantee technique by Li et al. [172] trade off privacy for a greater accuracy.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [172]
Not checked in the free run.
- Premisereasoning
Network communication is much slower than local computation in federated networks with many devices.
“There can be a large number of devices in a federated network, which means that network communication is much slower than local computation.”
Quote in the paperSubstantial claim - Premisereasoning
In peer-to-peer FL architecture, there is no central server for model aggregation.
“As illustrated in Fig. 2, there is no concept of a central server in peer-to-peer architecture like in the clientserver architecture for model aggregations.”
Quote in the paperNot a substantial claim - Interpretationreasoning
Statistical heterogeneity issues arise due to violation of IID assumptions in distributed optimization.
“statistical heterogeneity issues arise due to violation of independent and identically distributed (IID) assumptions in distributed optimization.”
Quote in the paperSubstantial claim- Rests on an assumption: Different devices across the network often have nonidentically distributed data.
- Rests on an assumption: The number of data points across devices also vary.
- Interpretationcitation
Peer-to-peer FL architecture is more secure because participating clients communicate directly without a third-party coordinator.
“Such architecture is more secure since the participating clients communicate directly without a third-party coordinator [127].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [127] source
Not checked in the free run.
- Premisereasoning
Expensive communication is a major challenge in FL systems.
“Expensive communication is a major challenge in FL systems.”
Quote in the paperSubstantial claim - Interpretationcitation
For moderate, naturally arising heterogeneity, adaptive optimizer is quite effective, especially in cross-device setting.
“But for moderate, naturally arising heterogeneity, adaptive optimizer is quite effective, especially in cross-device setting.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [128] source
Not checked in the free run.
- Interpretationreasoning
Clients for whom the global model does not perform as well as local models are disincentivized to take part in the federated network.
“So, they are disincentivized to take part in the federated network.”
Quote in the paperNot a substantial claim
- Classified as probably not a substantial claim. 41
- May combine two claims. 8
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationAdaptive Federated Optimization introduces adaptability in server optimization.
“Adaptive Federated Optimization [128] proposed by Google Research team introduces the adaptability in server optimization.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [128] source
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Findingnone givenFedAvg is robust to unbalanced and non-IID data distribution.
“It is robust to unbalanced and non-IID data distribution.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationFedAvg is based on an SGD optimization algorithm.
“The Federated averaging (FedAvg) algorithm [1] proposed by Google is based on an Stochastic Gradient Descent (SGD) optimization algorithm.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [1]
Not checked in the free run.
May combine two claims.
FindingcitationScaffold helps overcome client heterogeneity and reduce communication rounds in model convergence.
“This strategy helps to overcome the problem of client heterogeneity and reduce the communication round in model convergence.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [58]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationMini-batch optimization methods have proven successful in many cases.
“As an extension of classical stochastic methods, mini-batch optimization methods have proven to be successful in many cases [136].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [136]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationArivazhagan et al. propose a neural network architecture where global data is used to train only the base layers, while the personalization layers are trained on local data.
“Arivazhagan et al. [134] propose a neural network architecture, where global data is used to train only the base layers, while the personalization layers are trained on local data.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [134] source
Not checked in the free run.
May combine two claims.
FindingcitationBased on similarity of clients' data, different groups can be formed and a different model can be trained for each similar cluster.
“Moreover, based on similarity of clients' data, different groups can be formed and a different model can be trained for each similar cluster [97].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [97] source
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationHanzely et al. propose a new gradient descent variant called LLGD that allows each device to learn a mixture of its own local model and the global model.
“A new gradient descent variant called Loopless Gradient Descent (LLGD) by Hanzely et al. [135] allow each device to learn a mixture of its own local model and the global model.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [135]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationWu et al. mention three major challenges handled by FL process during personalization: device heterogeneity, data heterogeneity, and model heterogeneity.
“Wu et al. [105] mention three major challenges handled by FL process during personalization.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [105]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationSecure Aggregation is fault-tolerant up to 1/3rd of users.
“It is fault-tolerant up to 1/3rd of users, i.e., it works well even if 1/3rd of the clients fails to engage in the aggregation.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [19]
Not checked in the free run.
May combine two claims.
FindingcitationFedProx addresses system heterogeneity and statistical heterogeneity.
“FedProx [4] addresses the two inherent challenges of FL.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [4]
Not checked in the free run.
May combine two claims.
FindingcitationSecure Aggregation does not share any information with each other except the learnable parameters derived from aggregation, thus defending the privacy of each client's model.
“It does not share any information with each other except the learnable parameters derived from aggregation and thus defends the privacy of each client's model.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [19]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationRecent works like [6] have carried out decentralized training over heterogeneous data.
“Recent works like [6] have carried out decentralized training over heterogeneous data.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [6]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationFedProx is a re-parametrized and generalized version of FedAvg.
“It is a re-parametrized and generalized version of FedAvg.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [4]
Not checked in the free run.
Probably not a substantial claim: may not pass S4 (based). May combine two claims.
Findingnone givenWireless connectivity might not be available consistently, and as such many participating devices may drop from the FL system during training.
“However, wireless connectivity might not be available consistently, and as such many participating devices may drop from the FL system during training.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationThe authors of [146] have developed tools to measure statistical heterogeneity using metrics such as local dissimilarity.
“The authors of [146] have developed tools to measure statistical heterogeneity using metrics such as local dissimilarity.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [146] source
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
Findingnone givenSMC is one of the privacy mechanisms used in FL.
“Secure Multiparty Computation (SMC) is one of privacy mechanisms used in FL.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
FindingcitationData poisoning attacks may occur as clean-label attacks or dirty-label attacks.
“Data poisoning attacks may occur in two ways such as clean-label attack (adversary can poison correct class of data sample) and dirty-label attack (adversary try to misclassify the target label of training dataset) [147], [148].”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [147]
Not checked in the free run.
- cites [148] source
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationAccording to the survey of threat model by Lyu et al., inference attack falls into two categories: white-box attack and black-box attack.
“According to the survey of threat model by Lyu et al. [101], the inference attack falls into two categorieswhite-box attack and black-box attack.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [101]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationFault Tolerance is used because learning over remote devices becomes more critical as some devices often drop out even before an iteration is completed.
“Fault Tolerance [99] is used because learning over remote devices becomes more critical as some devices in the network often drop out even before an iteration is completed.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [99] source
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Findingnone givenThere can be significant variance in terms of the amount of data per device.
“Moreover, there can be significant variance in terms of the amount of data per device.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Findingnone givenIn cross-device FL, often only a few devices are active during an iteration.
“In cross-device FL, often only a few devices are active during an iteration.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationBonawitz et al. designed a communication-efficient SMC protocol for high-dimensional data to protect the privacy of users' model gradients.
“Bonawitz et al. designed a communication-efficient SMC protocol for highdimensional data to protect the privacy of users' model gradients [19]”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [19]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationAccording to Bhagoji et al., model poisoning is accomplished by an adversary controlling few malicious representatives with the aim to misclassify specific inputs with high confidence.
“According to Bhagoji et al. [55], model poisoning is accomplished by an adversary controlling few malicious representatives with the aim to misclassify specific inputs with high confidence.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [55] source
Not checked in the free run.
Classified as probably not a substantial claim.
FindingcitationThere are several open issues and challenges in FL.
“There are several open issues and challenges in FL [168].”
Quote in the paperSubstantial claim0 of 1 sources checked- cites [168]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationWei et al. proposed an aggregation algorithm called NbAFL where they add noise to client-side parameters before aggregation.
“Wei et al. [156] proposed an aggregation algorithm called NbAFL where they add noise to client-side parameters before aggregation.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [156]
Not checked in the free run.
Probably not a substantial claim: may not pass S4 (based).
Findingnone givenThere is a tradeoff between communication cost and accuracy in FL.
“There is a tradeoff between communication cost and accuracy in FL.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based).
Findingnone givenRecent research demonstrated that FL may not always provide sufficient privacy guarantee during model update.
“But recent research demonstrated that FL may not always provide sufficient privacy guarantee during model update.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationXu et al. proposed a double-masking protocol to guarantee the privacy of users' local gradients during FL.
“In order to guarantee the privacy of users' local gradients during FL, Xu et al. [114] proposed a double-masking protocol.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [114]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
FindingcitationGeyer et al. developed a method to obtain differential privacy at client level for FL.
“Geyer et al. [4] developed a method to obtain differential privacy at client level for FL.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [4]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
Findingnone givenHE is another security mechanism in FL.
“Homomorphic Encryption (HE) is another security mechanism in FL.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Findingnone givenFL has mainly been applied to supervised learning problems.
“FL has mainly been applied to supervised learning problems.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based).
Findingnone givenThere has been a growing interest in FL research in both industry and academia in recent years.
“There has been a growing interest in FL research in both industry and academia in recent years.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based). May combine two claims.
Findingnone givenThe challenges faced in the implementation of FL solutions for different application areas have not yet been thoroughly studied, with the focus of current studies being primarily on the training of FL models.
“The challenges faced in the implementation of FL solutions for different application areas have not yet been thoroughly studied, with the focus of current studies being primarily on the training of FL models.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Interpretationnone givenThe client-server architecture incurs less communication overhead.
“The client-server architecture incurs less communication overhead.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
InterpretationreasoningPeer-to-peer architecture is more secure since participating clients communicate directly without a third-party coordinator.
“the peer-to-peer architecture is even more secure since the participating clients communicate directly without a third-party coordinator.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Interpretationnone givenFedAvg is the best fit for HFL with a client-server architecture.
“This aggregation algorithm is the best fit for HFL with a client-server architecture.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationFTL gained enormous attention in various industries, especially in the healthcare sector.
“FTL gained enormous attention in various industries, especially in healthcare sector [121].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [121]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Interpretationnone givenPeer-to-peer architecture requires more computation for message encryption and decryption.
“peer-to-peer architecture requires more computation for message encryption and decryption.”
Quote in the paperNot a substantial claimMay combine two claims.
PremisecitationBoth the actual sensitive data and models are preserved locally in FTL.
“both the actual sensitive data and models are preserved locally [122].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [122]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationFTL was first proposed in [71].
“FTL was first proposed in [71].”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites [71] source
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationFL was first introduced by researchers at Google to update language models in Google's keyboard system for word autocompletion.
“FL was first introduced by the researchers at Google to update language models [1], [2] in Google's keyboard system for word autocompletion.”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites [1]
Not checked in the free run.
- cites [2]
Not checked in the free run.
Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Interpretationnone givenAdding contextual features to datasets in a privacy-preserving manner can lead to predictions that are more personalized.
“Adding contextual features to datasets in a privacypreserving manner can lead to predictions that are more personalized.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing), S4 (based).
Premisenone givenImportant-based updating utilizes the fact that most parameter values of a deep neural network model are sparsely distributed.
“Important-based updating utilizes the fact that most parameter values of a deep neural network model are sparsely distributed.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based).
Premisenone givenFL offers an emerging paradigm for facilitating multiple organization data collaboration without revealing their private data to each other.
“FL offers an emerging paradigm for facilitating multiple organization data collaboration without revealing their private data to each other.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S4 (based).
InterpretationreasoningFL enables certain industries like healthcare to overcome challenges related to data collection and privacy.
“FL enables certain industries like healthcare to overcome challenges related to data collection and privacy.”
Quote in the paperNot a substantial claim- Rests on an assumption: FL allows participating organizations to collaboratively train prediction models without having to share their data.
The paper relies on this without support.
Horizontal and Vertical FL can be ineffective when the data is highly heterogeneous.
“Horizontal and Vertical FL can be ineffective when the data is highly heterogeneous.”
The paper relies on this without support.
FTL is an effective approach when data is highly heterogeneous.
“In such cases, FTL is an effective approach where it transfers knowledge learned from source domain to a target domain.”
The paper relies on this without support.
Only a few selected devices may participate during a training iteration, and some devices may drop out due to connectivity or power issues.
“Due to system and network constraints in a number of settings, only a few selected devices may participate during a training iteration, and some devices may even drop out during an iteration due to connectivity or power issues.”
The paper relies on this without support.
Different devices across the network often have nonidentically distributed data.
“The violation occurs because different devices across the network often have nonidentically distributed data.”
The paper relies on this without support.
The number of data points across devices also vary.
“The number of data points across devices also vary.”
The paper relies on this without support.
Privacy/security issues are at the core of FL applications.
“privacy/security issues are at the core of FL applications.”
The paper relies on this without support.
Increased privacy/security using novel methods often comes at the cost of decreased system efficiency or model performance.
“Increased privacy/security using novel methods often comes at the cost of decreased system efficiency or model performance.”
The paper relies on this without support.
Currently, almost all implementations of FL use client-server architecture.
“Currently, almost all implementation of FL use client-server architecture.”
The paper relies on this without support.
Collaborative training in FL can result in better models compared to models trained by individual organizations or devices.
“In addition to the privacy and security benefits, collaborative training in FL can result in better models compared to models trained by individual organizations or devices.”
The paper relies on this without support.
Added security and privacy results in some accuracy loss in FL.
“In practice, this added security and privacy results in some accuracy loss, but it is often worth it for specific application domains.”
The paper relies on this without support.
FL allows participating organizations to collaboratively train prediction models without having to share their data.
“FL allows participating organizations to collaboratively train prediction models without having to share their data.”
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
FL does not cover all potential privacy threats.
“Generally, FL is invented to fulfill the privacy concern of private data, unfortunately it does not cover all potential privacy threats [93].”
From A systematic review on federated learning in medical image analysis (2023) · cites this paper as [93]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Hybrid-FL extension is assigned to the wrong reference
MinorThe systems-heterogeneity discussion attributes a FedCS extension called Hybrid-FL to [20], but reference [20] concerns private distributed aggregation. The Hybrid-FL paper is listed at [34].
“The FedCS protocol is extended by the authors in [20].Their Hybrid-FL protocol addresses the differences that exists in data distributions of participating clients.”
“20. B. Ghazi, R. Pagh, and A. Velingker, "Scalable and differentially private distributed aggregation in the shuffled model," arXiv. 2019.”
Communication table misattributes Reisizadeh et al.
MinorTable V assigns [131] to Reisizadeh et al.'s periodically averaged, quantized uploads. Reference [131] is instead a meta-learning paper by Finn et al.
“Reisizadeh et al. [131] Local Updating Method with models periodically averaged at the server and quantized uploads from edge devices.”
“131. C. Finn, P. Abbeel, and S. Levine, "Model-agnostic meta- learning for fast adaptation of deep networks," 2017.”
Threat-model table misattributes Shafahi et al.
MinorTable VII assigns [148] to Shafahi et al.'s poisoning work, but reference [148] is a BadNets paper by Gu et al. The Shafahi paper is listed at [147].
“Shafahi et al. [148] Poisoning Attack Data Poisoning "Watermarking" strategy (frog image) Data”
“148. T. Gu, B. Dolan-Gavitt, and S. Garg, "Badnets: Identifying vulnerabilities in the machine learning model supply chain," arXiv. 2017.”
FedProx has conflicting citations
MinorTable III cites FedProx as [15], while the algorithm discussion cites it as [4]. Reference [15] is about coded computation over heterogeneous clusters.
“FedBoost [129] FedProx [15] FedMA [110] Secure Aggregation [30]”
“5) FedProx [4] addresses the two inherent challenges of FL.”
Scaffold has conflicting citations
MinorThe discussion identifies Scaffold as [58], but Table III labels it [67]. Reference [67] is about association-rule mining, not Scaffold.
“FedAvg [1] Scaffold [67] Adaptive Federated Optimization [128] FedBoost [129]”
“2) Scaffold (Stochastic Controlled Averaging Federated Learning) [58] solves the problem of client drifting using variance reduction technique in its local update.”
Personalization table misattributes reference [100]
MinorTable IV names Mansour et al. for [100], but reference [100] is a meta-learning paper by Fallah et al. Mansour et al. appear at [97].
“Mansour et al. [100] Clustering”
“100. A. Fallah, A. Mokhtari, and A. Ozdaglar, "Personalized federated learning: A meta-learning approach," arXiv. 2020.”
Communication table misattributes Rothchild et al.
MinorTable V assigns [130] to Rothchild et al.'s compression work, but reference [130] is a personalization-evaluation paper by Wang et al. Rothchild et al.'s FetchSGD paper is listed at [141].
“Rothchild et al. [130] Compression A Count Sketch is used to compress the updates of model.Then, leverages sketch mergeability.”
“130. K. Wang, C. Kiddon, H. Eichner, and D. Ramage, "Federated Evaluation of On-device Personalization," 2018.”
Geyer et al. are assigned the FedProx reference
MinorThe privacy discussion credits Geyer et al.'s client-level differential-privacy method to [4], but reference [4] is the Li et al. FedProx paper. Geyer et al.'s method is listed at [23].
“Geyer et al. [4] developed a method to obtain differential privacy at client level for FL.”
“23. R. C. Geyer, T. Klein, and M. Nabi, "Differentially private federated learning: A client level perspective," arXiv. 2017.”
Briggs survey is cited as an unrelated paper
MinorTable I assigns [5] to the Briggs et al. IoT survey, whereas reference [5] is a communication-compression paper. The Briggs survey appears at [57].
“Briggs et al. [5] Focusing on IoT, covers works related to FL challenges and privacy preserving methods, identify the strengths and weaknesses of different methods applied to FL, and outlines future directions.”
“5. H. Tang, S. Gan, C. Zhang, T. Zhang, and J. Liu, "Communication compression for decentralized training," 2018.”
Systems-heterogeneity table misattributes FedCS
MinorTable VI names Nishio et al. and FedCS under [43], but reference [43] is by Feng et al. The Nishio and Yonetani client-selection paper is reference [33].
“Nishio et al. [43] Client Participation, concerned with number of clients FedCS, Federated Learning with client selection.”
“43. S. Feng, D. Niyato, P. Wang, D. I. Kim, and Y. C. Liang, "Joint service pricing and cooperative relay communication for federated learning," 2019, doi: 10.1109/iThings/GreenCom/CPSCom/SmartData.2019.00148.”
Threat-model table misattributes Bagdasaryan et al.
MinorTable VII cites Bagdasaryan et al. as [151], whereas reference [151] is a paper by Melis et al. Bagdasaryan et al.'s backdoor paper is listed at [149].
“Bagdasaryan et al. [151] Backdoored image-classification model, backdoored wordprediction model”
“151. L. Melis, C. Song, E. De Cristofaro, and V. Shmatikov, "Exploiting unintended feature leakage in collaborative learning," 2019, doi: 10.1109/SP.2019.00029.”
Lim survey has conflicting reference numbers
MinorTable I cites the Lim et al. mobile-edge-networks survey as [4], but reference [4] is a paper by Li et al. The Lim survey is listed as reference [25].
“Lim et al. [4] Highlights challenges of FL implementation and existing solutions and presents applications of FL for mobile edge network optimization.”
“4. T. Li, A. K. Sahu, M. Zaheer, M. Sanjabi, A. Talwalkar, and V. Smith, "Federated Optimization in Heterogeneous Networks," 2018, [Online]. Available: http://arxiv.org/abs/1812.06127.”
Personalization table misattributes reference [133]
MinorTable IV names McMahan et al. for [133], but reference [133] is a FedMD paper by Li and Wang.
“McMahan et al. [133] Adding user context Context featurized”
“133. D. Li and J. Wang, "FedMD: Heterogenous federated learning via model distillation," arXiv. 2019.”
Cites
Pan, S. J., & Yang, Q. (2010) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2009.191 · Cited by 23,387 · see citing works
Communication-Efficient Learning of Deep Networks from Decentralized Data
McMahan, H. B., Moore, E., Ramage, D., Hampson, S., & Arcas, B. A. Y. (2016) · arXiv · doi:10.48550/arxiv.1602.05629 · Cited by 14,215 · see citing works
Shamir, A. (1979) · Communications of the ACM · doi:10.1145/359168.359176 · Cited by 13,756 · see citing works
Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
Finn, C., Abbeel, P., & Levine, S. (2017) · arXiv · doi:10.48550/arxiv.1703.03400 · Cited by 8,512 · see citing works
Calibrating Noise to Sensitivity in Private Data Analysis
Dwork, C., McSherry, F., Nissim, K., & Smith, A. (2006) · Lecture Notes in Computer Science · doi:10.1007/11681878_14 · Cited by 7,272 · see citing works
Deep Learning with Differential Privacy
Abadi, M., Chu, A., Goodfellow, I., McMahan, H. B., Mironov, I., Talwar, K., & Zhang, L. (2016) · Proceedings of the 2016 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2976749.2978318 · Cited by 6,578 · see citing works
Federated Machine Learning: Concept and Applications
Yang, Q., Liu, Y., Chen, T., & Tong, Y. (2019) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3298981 · Cited by 6,322 · see citing works
Federated learning: Challenges, methods, and future directions
Li, T., Sahu, A. K., Talwalkar, A., & Smith, V. (2020) · IEEE Signal Processing Magazine · doi:10.1109/msp.2020.2975749 · Cited by 6,210 · see citing works
The Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2014) · Foundations and Trends® in Theoretical Computer Science · doi:10.1561/0400000042 · Cited by 5,557 · see citing works
Han, S., Mao, H., & Dally, W. J. (2015) · arXiv · doi:10.48550/arxiv.1510.00149 · Cited by 4,662 · see citing works
Federated Optimization in Heterogeneous Networks
Li, T., Sahu, A. K., Zaheer, M., Sanjabi, M., Talwalkar, A., & Smith, V. (2018) · arXiv · doi:10.48550/arxiv.1812.06127 · Cited by 4,023 · see citing works
Practical Secure Aggregation for Privacy-Preserving Machine Learning
Bonawitz, K., Ivanov, V., Kreuter, B., Marcedone, A., McMahan, H. B., Patel, S., Ramage, D., Segal, A., & Seth, K. (2017) · Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3133956.3133982 · Cited by 3,956 · see citing works
The Algorithmic Foundations of Differential Privacy
Dwork, C., & Roth, A. (2013) · doi:10.1561/9781601988195 · Cited by 3,856 · see citing works
Federated Learning: Strategies for Improving Communication Efficiency
Konečný, J., McMahan, H. B., Yu, F. X., Richtárik, P., Suresh, A. T., & Bacon, D. (2016) · arXiv · doi:10.48550/arxiv.1610.05492 · Cited by 3,766 · see citing works
Federated Learning in Mobile Edge Networks: A Comprehensive Survey
Lim, W. Y. B., Luong, N. C., Hoang, D. T., Jiao, Y., Liang, Y.-C., Yang, Q., Niyato, D., & Miao, C. (2020) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2020.2986024 · Cited by 2,565 · see citing works
Federated Learning With Differential Privacy: Algorithms and Performance Analysis
Wei, K., Li, J., Ding, M., Ma, C., Yang, H. H., Farokhi, F., Jin, S., Quek, T. Q. S., & Vincent Poor, H. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2020.2988575 · Cited by 2,400 · see citing works
Privacy-preserving deep learning
Shokri, R., & Shmatikov, V. (2015) · Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/2810103.2813687 · Cited by 2,356 · see citing works
Towards Federated Learning at Scale: System Design
Bonawitz, K., Eichner, H., Grieskamp, W., Huba, D., Ingerman, A., Ivanov, V., Kiddon, C., Konečný, J., Mazzocchi, S., McMahan, H. B., et al. (2019) · arXiv · doi:10.48550/arxiv.1902.01046 · Cited by 2,334 · see citing works
Adaptive Federated Learning in Resource Constrained Edge Computing Systems
Wang, S., Tuor, T., Salonidis, T., Leung, K. K., Makaya, C., He, T., & Chan, K. (2019) · IEEE Journal on Selected Areas in Communications · doi:10.1109/jsac.2019.2904348 · Cited by 2,232 · see citing works
Zhu, L., & Han, S. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-63076-8_2 · Cited by 1,906 · see citing works
(1978) · Cited by 1,875 · see citing works
Client Selection for Federated Learning with Heterogeneous Resources in Mobile Edge
Nishio, T., & Yonetani, R. (2019) · ICC 2019 - 2019 IEEE International Conference on Communications (ICC) · doi:10.1109/icc.2019.8761315 · Cited by 1,875 · see citing works
Exploiting Unintended Feature Leakage in Collaborative Learning
Melis, L., Song, C., De Cristofaro, E., & Shmatikov, V. (2019) · 2019 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2019.00029 · Cited by 1,734 · see citing works
SCAFFOLD: Stochastic Controlled Averaging for Federated Learning
Karimireddy, S. P., Kale, S., Mohri, M., Reddi, S. J., Stich, S. U., & Suresh, A. T. (2019) · arXiv · doi:10.48550/arxiv.1910.06378 · Cited by 1,734 · see citing works
Privacy-Preserving Deep Learning via Additively Homomorphic Encryption
Phong, L. T., Aono, Y., Hayashi, T., Wang, L., & Moriai, S. (2018) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2017.2787987 · Cited by 1,662 · see citing works
A review of applications in federated learning
Li, L., Fan, Y., Tse, M., & Lin, K.-Y. (2020) · Computers & Industrial Engineering · doi:10.1016/j.cie.2020.106854 · Cited by 1,559 · see citing works
Federated Learning for Healthcare Informatics
Xu, J., Glicksberg, B. S., Su, C., Walker, P., Bian, J., & Wang, F. (2021) · Journal of Healthcare Informatics Research · doi:10.1007/s41666-020-00082-4 · Cited by 1,542 · cited 1× · see citing works
“The challenges and approaches discussed in [62] were centered around the healthcare domain.”
BackgroundHow to backdoor federated learning
Bagdasaryan, E., Veit, A., Hua, Y., Estrin, D., & Shmatikov, V. (2018) · arXiv · doi:10.48550/arxiv.1807.00459 · Cited by 1,495 · see citing works
A survey on security and privacy of federated learning
Mothukuri, V., Parizi, R. M., Pouriyeh, S., Huang, Y., Dehghantanha, A., & Srivastava, G. (2021) · Future Generation Computer Systems · doi:10.1016/j.future.2020.10.007 · Cited by 1,480 · see citing works
A Survey on Federated Learning Systems: Vision, Hype and Reality for Data Privacy and Protection
Li, Q., Wen, Z., Wu, Z., Hu, S., Wang, N., Li, Y., Liu, X., & He, B. (2023) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3124599 · Cited by 1,407 · see citing works
Deep Models Under the GAN: Information Leakage from Collaborative Deep Learning
Hitaj, B., Ateniese, G., & Perez-Cruz, F. (2017) · Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3133956.3134012 · Cited by 1,377 · see citing works
Blockchain and Federated Learning for Privacy-Preserved Data Sharing in Industrial IoT
Lu, Y., Huang, X., Dai, Y., Maharjan, S., & Zhang, Y. (2020) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2019.2942190 · Cited by 1,245 · see citing works
Federated Learning for Mobile Keyboard Prediction
Hard, A., Rao, K., Mathews, R., Ramaswamy, S., Beaufays, F., Augenstein, S., Eichner, H., Kiddon, C., & Ramage, D. (2018) · arXiv · doi:10.48550/arxiv.1811.03604 · Cited by 1,148 · see citing works
Deep Gradient Compression: Reducing the Communication Bandwidth for Distributed Training
Lin, Y., Han, S., Mao, H., Wang, Y., & Dally, W. J. (2017) · arXiv · doi:10.48550/arxiv.1712.01887 · Cited by 1,134 · see citing works
Parallelized Stochastic Gradient Descent
Zinkevich, M., Weimer, M., Li, L., & Smola, A. J. (2010) · Cited by 1,124 · see citing works
Christen, P. (2012) · Springer eBooks · doi:10.1007/978-3-642-31164-2 · Cited by 1,111 · see citing works
Federated Learning via Over-the-Air Computation
Yang, K., Jiang, T., Shi, Y., & Ding, Z. (2020) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2019.2961673 · Cited by 1,084 · see citing works
BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain
Gu, T., Dolan-Gavitt, B., & Garg, S. (2017) · arXiv · doi:10.48550/arxiv.1708.06733 · Cited by 1,061 · see citing works
Differentially Private Federated Learning: A Client Level Perspective
Geyer, R. C., Klein, T., & Nabi, M. (2017) · arXiv · doi:10.48550/arxiv.1712.07557 · Cited by 1,052 · see citing works
In-Edge AI: Intelligentizing Mobile Edge Computing, Caching and Communication by Federated Learning
Wang, X., Han, Y., Wang, C., Zhao, Q., Chen, X., & Chen, M. (2019) · IEEE Network · doi:10.1109/mnet.2019.1800286 · Cited by 1,030 · cited 1× · see citing works
“Applications of FL in mobile devices included predicting user input [74], [86], [87], emoji [88], human trajectory [89], and human behavior [90]; reducing network congestion [91]; detecting physical hazards (smart-home IoT) [92]; performing industrial engineering—environmental”
BackgroundA Hybrid Approach to Privacy-Preserving Federated Learning
Truex, S., Baracaldo, N., Anwar, A., Steinke, T., Ludwig, H., Zhang, R., & Zhou, Y. (2019) · Proceedings of the 12th ACM Workshop on Artificial Intelligence and Security · doi:10.1145/3338501.3357370 · Cited by 1,029 · see citing works
FedHealth: A Federated Transfer Learning Framework for Wearable Healthcare
Chen, Y., Qin, X., Wang, J., Yu, C., & Gao, W. (2020) · IEEE Intelligent Systems · doi:10.1109/mis.2020.2988604 · Cited by 1,005 · see citing works
Privacy preserving association rule mining in vertically partitioned data
Vaidya, J., & Clifton, C. (2002) · Proceedings of the eighth ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/775047.775142 · Cited by 1,001 · see citing works
Federated learning of predictive models from federated Electronic Health Records
Brisimi, T. S., Chen, R., Mela, T., Olshevsky, A., Paschalidis, I. C., & Shi, W. (2018) · International Journal of Medical Informatics · doi:10.1016/j.ijmedinf.2018.01.007 · Cited by 997 · cited 2× · see citing works
“Data partitioning schemes, namely HFL [74], VFL [75]–[78], and hybrid FL [79], [80], as well as centralized [11], [81] and decentralized [82] designs for communication architecture and cross-silo [83] versus cross-device [84], [85] FL for the federation scale, are the other challenges and…”
BackgroundMethodKang, J., Xiong, Z., Niyato, D. T., Xie, S., & Zhang, J. (2019) · IEEE Internet of Things Journal · doi:10.1109/jiot.2019.2940820 · Cited by 961 · see citing works
Federated Learning of Deep Networks using Model AveragingPreprint
McMahan, H. B., Moore, E., Ramage, D., & y Arcas, B. A. (2016) · arXiv (Cornell University) · Cited by 923 · see citing works
Learning Differentially Private Recurrent Language Models
McMahan, H. B., Ramage, D., Talwar, K., & Zhang, L. (2017) · arXiv · doi:10.48550/arxiv.1710.06963 · Cited by 849 · see citing works
Adaptive Federated Optimization
Reddi, S., Charles, Z., Zaheer, M., Garrett, Z., Rush, K., Konečný, J., Kumar, S., & McMahan, H. B. (2020) · arXiv · doi:10.48550/arxiv.2003.00295 · Cited by 840 · see citing works
Verifynet: Secure and verifiable federated learning
Xu, G., Li, H., Liu, S., Yang, K., & Lin, X. (2020) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2019.2929409 · Cited by 839 · see citing works
Speeding Up Distributed Machine Learning Using Codes
Lee, K., Lam, M., Pedarsani, R., Papailiopoulos, D., & Ramchandran, K. (2018) · IEEE Transactions on Information Theory · doi:10.1109/tit.2017.2736066 · Cited by 829 · see citing works
Broadband Analog Aggregation for Low-Latency Federated Edge Learning
Zhu, G., Wang, Y., & Huang, K. (2019) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2019.2946245 · Cited by 786 · see citing works
Personalized Federated Learning: A Meta-Learning ApproachPreprint
Fallah, A., Mokhtari, A., & Ozdaglar, A. (2020) · arXiv.org · doi:10.48550/arxiv.2002.07948 · Cited by 751 · see citing works
Federated Learning: A Survey on Enabling Technologies, Protocols, and Applications
Aledhari, M., Razzak, R., Parizi, R. M., & Saeed, F. (2020) · IEEE Access · doi:10.1109/access.2020.3013541 · Cited by 733 · see citing works
Analyzing Federated Learning through an Adversarial Lens
Bhagoji, A. N., Chakraborty, S., Mittal, P., & Calo, S. (2018) · arXiv · doi:10.48550/arxiv.1811.12470 · Cited by 703 · see citing works
Sharemind: A Framework for Fast Privacy-Preserving Computations
Bogdanov, D., Laur, S., & Willemson, J. (2008) · Lecture notes in computer science · doi:10.1007/978-3-540-88313-5_13 · Cited by 698 · see citing works
SecureBoost: A Lossless Federated Learning Framework
Cheng, K., Fan, T., Jin, Y., Liu, Y., Chen, T., Papadopoulos, D., & Yang, Q. (2021) · IEEE Intelligent Systems · doi:10.1109/mis.2021.3082561 · Cited by 669 · see citing works
SecureBoost: A Lossless Federated Learning Framework
Cheng, K., Fan, T., Jin, Y., Liu, Y., Chen, T., Papadopoulos, D., & Yang, Q. (2019) · arXiv · doi:10.48550/arxiv.1901.08755 · Cited by 668 · see citing works
A secure federated transfer learning framework
Liu, Y., Kang, Y., Xing, C., Chen, T., & Yang, Q. (2020) · IEEE Intelligent Systems · doi:10.1109/mis.2020.2988525 · Cited by 645 · see citing works
Privacy-preserving <i>k</i> -means clustering over vertically partitioned data
Vaidya, J., & Clifton, C. (2003) · Proceedings of the ninth ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/956750.956776 · Cited by 626 · see citing works
Federated Learning Over Wireless Fading Channels
Amiri, M. M., & Gunduz, D. (2020) · IEEE Transactions on Wireless Communications · doi:10.1109/twc.2020.2974748 · Cited by 625 · see citing works
Agnostic Federated LearningPreprint
Mohri, M., Sivek, G., & Suresh, A. T. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1902.00146 · Cited by 619 · see citing works
A Learning-Based Incentive Mechanism for Federated Learning
Zhan, Y., Li, P., Qu, Z., Zeng, D., & Guo, S. (2020) · IEEE Internet of Things Journal · doi:10.1109/jiot.2020.2967772 · Cited by 602 · see citing works
Federated Learning with Personalization LayersPreprint
Arivazhagan, M. G., Aggarwal, V., Singh, A. K., & Choudhary, S. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.00818 · Cited by 595 · see citing works
FedPAQ: A Communication-Efficient Federated Learning Method with Periodic Averaging and Quantization
Reisizadeh, A., Mokhtari, A., Hassani, H., Jadbabaie, A., & Pedarsani, R. (2019) · arXiv · doi:10.48550/arxiv.1909.13014 · Cited by 587 · see citing works
More Effective Distributed ML via a Stale Synchronous Parallel Parameter Server
Ho, Q., Cipar, J., Cui, H., Kim, J. K., Lee, S., Gibbons, P. B., Gibson, G. A., Ganger, G. R., & Xing, E. P. (2013) · PubMed · doi:10.1184/r1/6475898 · Cited by 586 · see citing works
Chen, Y., Sun, X., & Jin, Y. (2019) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2019.2953131 · Cited by 575 · see citing works
iDLG: Improved Deep Leakage from GradientsPreprint
Zhao, B., Mopuri, K. R., & Bilen, H. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2001.02610 · Cited by 575 · see citing works
FedMD: Heterogenous Federated Learning via Model DistillationPreprint
Li, D., & Wang, J. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1910.03581 · Cited by 562 · see citing works
Federated Learning for Edge Networks: Resource Optimization and Incentive Mechanism
Khan, L. U., Pandey, S. R., Tran, N. H., Saad, W., Han, Z., Nguyen, M. N. H., & Hong, C. S. (2020) · IEEE Communications Magazine · doi:10.1109/mcom.001.1900649 · Cited by 506 · see citing works
CMFL: Mitigating Communication Overhead for Federated Learning
Wang, L., Wang, W., & Li, B. (2019) · IEEE International Conference on Distributed Computing Systems (ICDCS) · doi:10.1109/icdcs.2019.00099 · Cited by 492 · see citing works
Fair Resource Allocation in Federated Learning
Li, T., Sanjabi, M., Beirami, A., & Smith, V. (2019) · arXiv · doi:10.48550/arxiv.1905.10497 · Cited by 475 · see citing works
Asynchronous Online Federated Learning for Edge Devices with Non-IID Data
Chen, Y., Ning, Y., Slawski, M., & Rangwala, H. (2020) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata50022.2020.9378161 · Cited by 475 · see citing works
Jeong, E., Oh, S., Kim, H., Park, J., Bennis, M., & Kim, S.-L. (2018) · arXiv · doi:10.48550/arxiv.1811.11479 · Cited by 465 · see citing works
Lyu, L., Yu, H., Zhao, J., & Yang, Q. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-63076-8_1 · Cited by 444 · see citing works
Gradient Coding: Avoiding Stragglers in Distributed Learning
Tandon, R., Lei, Q., Dimakis, A. G., & Karampatziakis, N. (2017) · International Conference on Machine Learning · Cited by 431 · see citing works
Hardy, S., Henecka, W., Ivey-Law, H., Nock, R., Patrini, G., Smith, G., & Thorne, B. (2017) · arXiv · doi:10.48550/arxiv.1711.10677 · Cited by 431 · see citing works
Personalized Federated Learning for Intelligent IoT Applications: A Cloud-Edge Based Framework
Wu, Q., He, K., & Chen, X. (2020) · IEEE Open Journal of the Computer Society · doi:10.1109/ojcs.2020.2993259 · Cited by 424 · see citing works
Calibrating Noise to Sensitivity in Private Data Analysis
Dwork, C., McSherry, F., Kobbi Nissim, B., & Smith, A. (2017) · Journal of Privacy and Confidentiality · doi:10.29012/jpc.v7i3.405 · Cited by 390 · see citing works
Mitigating Sybils in Federated Learning Poisoning
Fung, C., Yoon, C. J. M., & Beschastnikh, I. (2018) · arXiv · doi:10.48550/arxiv.1808.04866 · Cited by 385 · see citing works
Improving Federated Learning Personalization via Model Agnostic Meta Learning
Jiang, Y., Konečný, J., Rush, K., & Kannan, S. (2019) · arXiv · doi:10.48550/arxiv.1909.12488 · Cited by 381 · cited 1× · see citing works
“Moreover, as observed in [170], a gap exists between the accuracy of personalized and global models, making the case of personalization techniques an important research area in FL.”
BackgroundFederated Learning in Vehicular Edge Computing: A Selective Model Aggregation Approach
Ye, D., Yu, R., Pan, M., & Han, Z. (2020) · IEEE Access · doi:10.1109/access.2020.2968399 · Cited by 381 · see citing works
Survey of Personalization Techniques for Federated Learning
Kulkarni, V., Kulkarni, M., & Pant, A. (2020) · World Conference on Smart Trends in Systems, Security and Sustainability (WorldS4) · doi:10.1109/worlds450073.2020.9210355 · Cited by 381 · see citing works
Three Approaches for Personalization with Applications to Federated LearningPreprint
Mansour, Y., Mohri, M., Ro, J. H., & Suresh, A. T. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2002.10619 · Cited by 337 · see citing works
Xu, R., Baracaldo, N., Zhou, Y., Anwar, A., & Ludwig, H. (2019) · ACM Workshop on Artificial Intelligence and Security (AISec) · doi:10.1145/3338501.3357371 · Cited by 328 · see citing works
Lim, W. Y. B., Huang, J., Xiong, Z., Kang, J., Niyato, D. T., Hua, X., Leung, C., & Miao, C. (2021) · IEEE Transactions on Intelligent Transportation Systems · doi:10.1109/tits.2021.3056341 · Cited by 318 · see citing works
Expanding the Reach of Federated Learning by Reducing Client Resource Requirements
Caldas, S., Konečny, J., McMahan, H. B., & Talwalkar, A. (2018) · arXiv · doi:10.48550/arxiv.1812.07210 · Cited by 312 · see citing works
FedVision: An Online Visual Object Detection Platform Powered by Federated Learning
Liu, Y., Huang, A., Luo, Y., Huang, H., Liu, Y., Chen, Y., Feng, L., Chen, T., Yu, H., & Yang, Q. (2020) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v34i08.7021 · Cited by 310 · see citing works
Federated Learning for Keyword Spotting
Leroy, D., Coucke, A., Lavril, T., Gisselbrecht, T., & Dureau, J. (2019) · ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2019.8683546 · Cited by 291 · see citing works
Coded Computation Over Heterogeneous Clusters
Reisizadeh, A., Prakash, S., Pedarsani, R., & Avestimehr, A. S. (2017) · IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit.2017.8006961 · Cited by 269 · see citing works
Distributed Coordinate Descent Method for Learning with Big Data
Richtárik, P., & Takác, M. (2013) · Journal of machine learning research · doi:10.48550/arxiv.1310.2059 · Cited by 266 · see citing works
cpSGD: Communication-efficient and differentially-private distributed SGDPreprint
Agarwal, N., Suresh, A. T., Yu, F., Kumar, S., & McMahan, H. B. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.10559 · Cited by 264 · see citing works
Privacy Preserving Naïve Bayes Classifier for Vertically Partitioned Data
Vaidya, J. S., & Clifton, C. (2004) · SIAM International Conference on Data Mining · doi:10.1137/1.9781611972740.59 · Cited by 262 · see citing works
Protection Against Reconstruction and Its Applications in Private Federated LearningPreprint
Bhowmick, A., Duchi, J. C., Freudiger, J., Kapoor, G., & Rogers, R. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1812.00984 · Cited by 261 · see citing works
Incentive Design for Efficient Federated Learning in Mobile Networks: A Contract Theory Approach
Kang, J., Xiong, Z., Niyato, D., Yu, H., Liang, Y.-C., & Kim, D. I. (2019) · 2019 IEEE VTS Asia Pacific Wireless Communications Symposium (APWCS) · doi:10.1109/vts-apwcs.2019.8851649 · Cited by 255 · see citing works
Rouhani, B. D., Riazi, M. S., & Koushanfar, F. (2018) · Proceedings - ACM IEEE Design Automation Conference · doi:10.1145/3195970.3196023 · Cited by 254 · see citing works
DIAMETERS OF SOME FINITE-DIMENSIONAL SETS AND CLASSES OF SMOOTH FUNCTIONS
Kašin, B. S. (1977) · Mathematics of the USSR-Izvestiya · doi:10.1070/im1977v011n02abeh001719 · Cited by 251 · see citing works
Federated Learning of a Mixture of Global and Local ModelsPreprint
Hanzely, F., & Richtárik, P. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2002.05516 · Cited by 250 · see citing works
Knock Knock, Who's There? Membership Inference on Aggregate Location Data
Pyrgelis, A., Troncoso, C., & Cristofaro, E. D. (2018) · Network and Distributed System Security Symposium · doi:10.14722/ndss.2018.23183 · Cited by 232 · see citing works
Federated Learning for Emoji Prediction in a Mobile Keyboard
Ramaswamy, S., Mathews, R., Rao, K., & Beaufays, F. (2019) · arXiv · doi:10.48550/arxiv.1906.04329 · Cited by 231 · see citing works
Federated Learning in Vehicular Networks
Elbir, A. M., Soner, B., Çöleri, S., Gündüz, D., & Bennis, M. (2022) · IEEE International Mediterranean Conference on Communications and Networking (MeditCom) · doi:10.1109/meditcom55741.2022.9928621 · Cited by 218 · see citing works
When Edge Meets Learning: Adaptive Control for Resource-Constrained Distributed Machine Learning
Wang, S., Tuor, T., Salonidis, T., Leung, K. K., Makaya, C., He, T., & Chan, K. (2018) · IEEE INFOCOM 2018 - IEEE Conference on Computer Communications · doi:10.1109/infocom.2018.8486403 · Cited by 209 · see citing works
Motivating Workers in Federated Learning: A Stackelberg Game Perspective
Sarikaya, Y., & Erçetin, Ö. (2019) · IEEE Networking Letters · doi:10.1109/lnet.2019.2947144 · Cited by 209 · see citing works
Communication Compression for Decentralized Training
Tang, H., Gan, S., Zhang, C., Zhang, T., & Liu, J. (2018) · arXiv · doi:10.48550/arxiv.1803.06443 · Cited by 206 · see citing works
Privacy-Preserving Patient Similarity Learning in a Federated Environment: Development and Analysis
Lee, J., Sun, J., Wang, F., Wang, S., Jun, C.-H., & Jiang, X. (2018) · JMIR Medical Informatics · doi:10.2196/medinform.7744 · Cited by 196 · see citing works
Privacy-preserving distributed linear regression on high-dimensional data
Gascón, A., Schoppmann, P., Balle, B., Raykova, M., Doerner, J., Zahur, S., & Evans, D. (2017) · Proceedings on Privacy Enhancing Technologies · doi:10.1515/popets-2017-0053 · Cited by 193 · see citing works
Analyzing Federated Learning through an Adversarial Lens
Bhagoji, A. N., Chakraborty, S., Mittal, P., & Calo, S. B. (2019) · International Conference on Machine Learning · Cited by 188 · see citing works
Federated Evaluation of On-device Personalization
Wang, K., Mathews, R., Kiddon, C., Eichner, H., Beaufays, F., & Ramage, D. (2019) · arXiv · doi:10.48550/arxiv.1910.10252 · Cited by 184 · see citing works
DeepSecure: Scalable Provably-Secure Deep Learning
Rouhani, B. D., Riazi, M. S., & Koushanfar, F. (2018) · Proceedings - ACM IEEE Design Automation Conference · doi:10.1109/dac.2018.8465894 · Cited by 179 · see citing works
Human Activity Recognition Using Federated Learning
Sozinov, K., Vlassov, V., & Girdzijauskas, S. (2018) · 2018 IEEE Intl Conf on Parallel & Distributed Processing with Applications, Ubiquitous Computing & Communications, Big Data & Cloud Computing, Social Computing & Networking, Sustainable Computing & Communications (ISPA/IUCC/BDCloud/SocialCom/SustainCom) · doi:10.1109/bdcloud.2018.00164 · Cited by 167 · see citing works
Privacy-preserving decision trees over vertically partitioned data
Vaidya, J. S., Clifton, C., Kantarcıoğlu, M., & Patterson, A. S. (2008) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/1409620.1409624 · Cited by 166 · see citing works
Decentralized Collaborative Learning of Personalized Models over NetworksPreprint
Vanhaesebrouck, P., Bellet, A., & Tommasi, M. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1610.05202 · Cited by 158 · see citing works
Feng, J., Rong, C., Sun, F., Guo, D., & Li, Y. (2020) · Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies · doi:10.1145/3381006 · Cited by 158 · see citing works
cpSGD: communication-efficient and differentially-private distributed SGD
Agarwal, N., Suresh, A. T., Yu, F., Kumar, S., & McMahan, H. B. (2018) · neural information processing systems · Cited by 141 · see citing works
One-Shot Federated LearningPreprint
Guha, N., Talwalkar, A., & Smith, V. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1902.11175 · Cited by 137 · see citing works
Collaborative Deep Learning in Fixed Topology NetworksPreprint
Jiang, Z., Balu, A., Hegde, C., & Sarkar, S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1706.07880 · Cited by 134 · see citing works
Asynchronous Federated Learning for Geospatial Applications
Sprague, M., Jalalirad, A., Scavuzzo, M., et al. (2019) · Communications in computer and information science · doi:10.1007/978-3-030-14880-5_2 · Cited by 133 · see citing works
Joint Service Pricing and Cooperative Relay Communication for Federated Learning
Feng, S., Niyato, D. T., Wang, P., Kim, D. I., & Liang, Y. (2019) · IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) · doi:10.1109/ithings/greencom/cpscom/smartdata.2019.00148 · Cited by 130 · see citing works
Abnormal client behavior detection in federated learning
Li, S., Cheng, Y., Liu, Y., Wang, W., & Chen, T. (2019) · arXiv · doi:10.48550/arxiv.1910.09933 · Cited by 122 · see citing works
Fair Resource Allocation in Federated Learning
Li, T., Sanjabi, M., Beirami, A., & Smith, V. (2020) · International Conference on Learning Representations · Cited by 120 · see citing works
Two-Stream Federated Learning: Reduce the Communication Costs
Yao, X., Huang, C., & Sun, L. (2018) · IEEE International Conference on Visual Communications and Image Processing (VCIP) · doi:10.1109/vcip.2018.8698609 · Cited by 118 · see citing works
Secure and Efficient Federated Transfer Learning
Sharma, S., Xing, C., Liu, Y., & Kang, Y. (2019) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata47090.2019.9006280 · Cited by 115 · see citing works
Anh, T. T., Luong, N. C., Niyato, D. T., Kim, D. I., & Wang, L. (2019) · IEEE Wireless Communications Letters · doi:10.1109/lwc.2019.2917133 · Cited by 114 · see citing works
High-Performance Distributed ML at Scale through Parameter Server Consistency Models
Dai, W., Kumar, A., Wei, J., Ho, Q., Gibson, G. A., & Xing, E. P. (2015) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v29i1.9195 · Cited by 109 · see citing works
InPrivate Digging: Enabling Tree-based Distributed Data Mining with Differential Privacy
Zhao, L., Ni, L., Hu, S., Chen, Y., Zhou, P., Xiao, F., & Wu, L. (2018) · IEEE INFOCOM 2018 - IEEE Conference on Computer Communications · doi:10.1109/infocom.2018.8486352 · Cited by 108 · see citing works
A Review of Privacy-Preserving Federated Learning for the Internet-of-Things
Briggs, C., Fan, Z., & Andras, P. (2021) · Studies in Computational Intelligence · doi:10.1007/978-3-030-70604-3_2 · Cited by 101 · see citing works
Federated Learning Of Out-Of-Vocabulary Words
Chen, M., Mathews, R., Ouyang, T., & Beaufays, F. (2019) · arXiv · doi:10.48550/arxiv.1903.10635 · Cited by 97 · see citing works
On-Device Federated Learning via Blockchain and its Latency Analysis.
Kim, H., Park, J., Bennis, M., & Kim, S. (2018) · Cited by 94 · see citing works
COLA: Decentralized Linear LearningPreprint
He, L., Bian, A., & Jaggi, M. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1808.04883 · Cited by 94 · see citing works
Boosting Privately: Federated Extreme Gradient Boosting for Mobile Crowdsensing
Liu, Y., Ma, Z., Liu, X., Ma, S., Nepal, S., Deng, R. H., & Ren, K. (2020) · 2020 IEEE 40th International Conference on Distributed Computing Systems (ICDCS) · doi:10.1109/icdcs47774.2020.00017 · Cited by 93 · see citing works
eSGD: Communication Efficient Distributed Deep Learning on the Edge
Tao, Z., & Li, Q. (2018) · Cited by 92 · see citing works
Mandal, K., & Gong, G. (2019) · Cloud Computing Security Workshop · doi:10.1145/3338466.3358926 · Cited by 91 · see citing works
Privacy-Preserving Ridge Regression with only Linearly-Homomorphic Encryption
Giacomelli, I., Jha, S., Jóye, M., Page, C. D., & Yoon, K. (2018) · Lecture notes in computer science · doi:10.1007/978-3-319-93387-0_13 · Cited by 89 · see citing works
Measuring statistical heterogeneity: The Pietra index
Eliazar, I. I., & Sokolov, I. M. (2010) · Physica A: Statistical Mechanics and its Applications · doi:10.1016/j.physa.2009.08.006 · Cited by 85 · see citing works
Learning Context-Aware Policies from Multiple Smart Homes via Federated Multi-Task Learning
Yu, T., Li, T., Sun, Y., Nanda, S., Smith, V., Sekar, V., & Seshan, S. (2020) · 2020 IEEE/ACM Fifth International Conference on Internet-of-Things Design and Implementation (IoTDI) · doi:10.1109/iotdi49375.2020.00017 · Cited by 85 · see citing works
Differentially Private Meta-Learning
Li, J., Khodak, M., Caldas, S., & Talwalkar, A. (2019) · arXiv · doi:10.48550/arxiv.1909.05830 · Cited by 82 · see citing works
Private, Yet Practical, Multiparty Deep Learning
Zhang, X., Ji, S., Wang, H., & Wang, T. (2017) · IEEE International Conference on Distributed Computing Systems (ICDCS) · doi:10.1109/icdcs.2017.215 · Cited by 77 · see citing works
Nguyen, H. T., Cong Luong, N., Zhao, J., Yuen, C., & Niyato, D. (2020) · 2020 IEEE 6th World Forum on Internet of Things (WF-IoT) · doi:10.1109/wf-iot48130.2020.9221089 · Cited by 75 · see citing works
Triastcyn, A., & Faltings, B. V. (2020) · IEEE Intelligent Systems · doi:10.1109/mis.2020.2993966 · Cited by 69 · see citing works
HHHFL: Hierarchical Heterogeneous Horizontal Federated Learning for ElectroencephalographyPreprint
Gao, D., Ju, C., Wei, X., Liu, Y., Chen, T., & Yang, Q. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1909.05784 · Cited by 68 · see citing works
Federated Region-Learning: An Edge Computing Based Framework for Urban Environment Sensing
Hu, B., Gao, Y., Liu, L., & Ma, H. (2018) · 2018 IEEE Global Communications Conference (GLOBECOM) · doi:10.1109/glocom.2018.8647649 · Cited by 67 · see citing works
Multi-Participant Multi-Class Vertical Federated Learning
Feng, S., & Yu, H. (2020) · arXiv · doi:10.48550/arxiv.2001.11154 · Cited by 66 · see citing works
Semi-Cyclic Stochastic Gradient Descent
Eichner, H., Koren, T., McMahan, H. B., Srebro, N., & Talwar, K. (2019) · arXiv · doi:10.48550/arxiv.1904.10120 · Cited by 66 · see citing works
Fadl Federated-autonomous deep learning for distributed electronic health record
Liu, D., Miller, T., Sayeed, R., & Mandl, K. D. (2018) · arXiv · doi:10.48550/arxiv.1811.11400 · Cited by 65 · see citing works
High-Performance Distributed ML at Scale through Parameter Server Consistency ModelsPreprint
Dai, W., Kumar, A., Wei, J., et al. (2014) · arXiv (Cornell University) · doi:10.48550/arxiv.1410.8043 · Cited by 65 · see citing works
Scalable and Differentially Private Distributed Aggregation in the Shuffled Model
Ghazi, B., Pagh, R., & Velingker, A. (2019) · arXiv · doi:10.48550/arxiv.1906.08320 · Cited by 62 · see citing works
LoAdaBoost:Loss-Based AdaBoost Federated Machine Learning on medical DataPreprint
Li, H., Yin, Y., Fu, Z., Zhang, S., Deng, H., & Liu, D. (2018) · arXiv (Cornell University) · Cited by 57 · see citing works
ELFISH: Resource-Aware Federated Learning on Heterogeneous Edge Devices.Preprint
Xu, Z., Zhao, Y., Xiong, J., Yang, J., & Chen, X. (2019) · arXiv (Cornell University) · Cited by 56 · see citing works
A Communication Efficient Collaborative Learning Framework for Distributed FeaturesPreprint
Liu, Y., Kang, Y., Zhang, X., Li, L., Cheng, Y. J., Chen, T., Hong, M., & Yang, Q. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.11187 · Cited by 53 · see citing works
Pretraining Federated Text Models for Next Word Prediction
Stremmel, J., & Singh, A. (2021) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-73103-8_34 · Cited by 53 · see citing works
Towards Mitigating Device Heterogeneity in Federated Learning via Adaptive Model Quantization
Abdelmoniem, A. M., & Canini, M. (2021) · doi:10.1145/3437984.3458839 · Cited by 52 · see citing works
Privacy-Preserving Tensor Factorization for Collaborative Health Data Analysis
Ma, J., Zhang, Q., Lou, J., Ho, J. C., Xiong, L., & Jiang, X. (2019) · Proceedings of the 28th ACM International Conference on Information and Knowledge Management · doi:10.1145/3357384.3357878 · Cited by 51 · see citing works
Edge-Assisted Hierarchical Federated Learning with Non-IID DataPreprint
Liu, L., Zhang, J., Song, S., & Letaief, K. B. (2019) · arXiv (Cornell University) · Cited by 47 · see citing works
Approximate Gradient Coding via Sparse Random GraphsPreprint
Charles, Z., Papailiopoulos, D., & Ellenberg, J. S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1711.06771 · Cited by 46 · see citing works
Personalization of Deep Learning
Schneider, J., & Vlachos, M. (2021) · Springer eBooks · doi:10.1007/978-3-658-32182-6_14 · Cited by 44 · see citing works
A Survey on Entity Alignment of Knowledge Base
Zhuang, Y., Li, G., & Jianhua, F. (2016) · Journal of Computer Research and Development · doi:10.7544/issn1000-1239.2016.20150661 · Cited by 40 · see citing works
Federated Learning for Ranking Browser History SuggestionsPreprint
Hartmann, F., Suh, S., Komarzewski, A., Smith, T. D., & Segall, I. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1911.11807 · Cited by 39 · see citing works
Privacy Regulation Aware Process Mapping in Geo-Distributed Cloud Data Centers
Zhou, A. C., Xiao, Y., Gong, Y., He, B., Zhai, J., & Mao, R. (2019) · IEEE Transactions on Parallel and Distributed Systems · doi:10.1109/tpds.2019.2896894 · Cited by 37 · see citing works
Hybrid-FL: Cooperative Learning Mechanism Using Non-IID Data in Wireless Networks.Preprint
Yoshida, N., Nishio, T., Morikura, M., Yamamoto, K., & Yonetani, R. (2019) · arXiv (Cornell University) · Cited by 37 · see citing works
Securing Distributed Gradient Descent in High Dimensional Statistical Learning
Su, L., & Xu, J. (2019) · ACM SIGMETRICS International Conference on Measurement and Modeling of Computer Systems (SIGMETRICS) · doi:10.1145/3309697.3331499 · Cited by 36 · see citing works
Visual Inspection with Federated Learning
Xu, H., Yu, H., & Gu, H. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-27272-2_5 · Cited by 36 · see citing works
Towards Effective Device-Aware Federated Learning
Anelli, V. W., Deldjoo, Y., Noia, T. D., & Ferrara, A. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-35166-3_34 · Cited by 35 · see citing works
Lim, W. Y. B., Huang, J., Xiong, Z., et al. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2004.03877 · Cited by 24 · see citing works
Gradient Coding Using the Stochastic Block Model
Charles, Z., & Papailiopoulos, D. (2018) · 2018 IEEE International Symposium on Information Theory (ISIT) · doi:10.1109/isit.2018.8437887 · Cited by 23 · see citing works
Quantifying the Performance of Federated Transfer LearningPreprint
Jing, Q., Wang, W., Zhang, J., Han, T., & Chen, K. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1912.12795 · Cited by 23 · see citing works
Decentralized Bayesian Learning over Graphs
Lalitha, A., Wang, X., Kilinc, O., Lu, Y., Javidi, T., & Koushanfar, F. (2019) · arXiv · doi:10.48550/arxiv.1905.10466 · Cited by 20 · see citing works
Survey on Federated Learning Towards Privacy Preserving AI
Kurupathi, S. R., & Maaß, W. (2020) · doi:10.5121/csit.2020.101120 · Cited by 19 · see citing works
Heterogeneity-Aware Federated Learning.Preprint
Yang, C., Wang, Q., Xu, M., et al. (2020) · arXiv (Cornell University) · Cited by 14 · see citing works
A Federated Learning Framework for Privacy-preserving and Parallel Training.Preprint
Cao, T.-D., Truong-Huu, T., Tran, H., & Khanh, T. V. (2020) · arXiv (Cornell University) · Cited by 12 · see citing works
Securing Distributed Gradient Descent in High Dimensional Statistical Learning
Su, L., & Xu, J. (2019) · ACM SIGMETRICS Performance Evaluation Review · doi:10.1145/3376930.3376983 · Cited by 6 · see citing works
Federated Face Anti-spoofing.Preprint
Shao, R., Perera, P., Yuen, P. C., et al. (2020) · arXiv (Cornell University) · Cited by 5 · see citing works
Model Aggregation via Good-Enough Model SpacesPreprint
Guha, N., & Smith, V. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1805.07782 · Cited by 3 · see citing works
FedBoost: Communication-efficient algorithms for federated learning
Hamer, J., Mohri, M., & Suresh, A. T. (2020) · International Conference on Machine Learning · Cited by 3 · see citing works
Applied Federated Learning: Improving Google Keyboard Query Suggestions
Yang, T., Andrew, G., Eichner, H., Sun, H., Li, W., Kong, N., Ramage, D., & Beaufays, F. (2018) · arXiv · doi:10.48550/arxiv.1812.02903
Flagged first, then by citations
Research works printed but not linked: 12
- kairouz, 2019, Advances and open problems in federated learning, arXiv 1912 04977
- shafahi, 2018, Poison frogs! Targeted clean-label poisoning attacks on neural networks, arXiv 1804 00792
- fang, 2019, Local model poisoning attacks to byzantine-robust federated learning, arXiv 1911 11815
- takabi, 2016, Privacy preserving multiparty machine learning with homomorphic encryption, Proc 29th Annu Conf Neural Inf Process Syst
- khodak, 2019, Adaptive gradient-based meta-learning methods, arXiv 1906 02717
- rothchild, 2020, FetchSGD: Communication-efficient federated learning with sketching, arXiv 2007 07682
- zhang, 2015, Deep learning with elastic averaging SGD, arXiv 1412 6651
- lin, 2018, Don’t use large mini-batches, use local SGD, arXiv 1808 07217
- smith, 2017, Federated multi-task learning, arXiv 1705 10467
- wang, 2020, Federated learning with matched averaging, arXiv 2002 06440
- 10.1109/TIT.2019.2904055 · DOI
- caldas, 2018, Federated kernelized multi-task learning, Proc SysML Conf, 1
Cited by
Sorted by citations
Federated learning for medical image analysis: A survey
Guan, H., Yap, P.-T., Bozoki, A., & Liu, M. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.110424 · Cited by 518 · see citing works
Rahman, A., Hossain, M. S., Muhammad, G., Kundu, D., Debnath, T., Rahman, M., Khan, M. S. I., Tiwari, P., & Band, S. S. (2023) · Cluster Computing · doi:10.1007/s10586-022-03658-4 · Cited by 399 · see citing works
A systematic review on federated learning in medical image analysis
Sohan, M. F., & Basalamah, A. (2023) · IEEE Access · doi:10.1109/access.2023.3260027 · Cited by 114 · see citing works
Reviewing Federated Machine Learning and Its Use in Diseases Prediction
Moshawrab, M., Adda, M., Bouzouane, A., Ibrahim, H., & Raad, A. (2023) · Sensors · doi:10.3390/s23042112 · Cited by 114 · see citing works
A survey for federated learning evaluations: Goals and measures
Chai, D., Wang, L., Yang, L., Zhang, J., Chen, K., & Yang, Q. (2024) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2024.3382002 · Cited by 64 · see citing works
Federated learning vulnerabilities, threats and defenses: A systematic review and future directions
Almutairi, S., & Barnawi, A. (2023) · Internet of Things · doi:10.1016/j.iot.2023.100947 · Cited by 35 · see citing works
- Authors
- Rahman, K. M. J., Ahmed, F., Akhter, N., Hasan, M., Amin, R., Aziz, K. E., Islam, A. K. M. M., Mukta, M. S. H., & Islam, A. K. M. N.
- Affiliations
- United International University (BD)
- ORCID
- Venue
- IEEE Access
- Published
- 2021
- Volume
- 9
- Pages
- 124682-124700
- ISSN
- 2169-3536 (ISSN-L)
- OpenAlex
- W3196965143
- Semantic Scholar
- 5e3a44f8f0b4267d79377d1f948500f65fe76726
- Cited by
- 147
- References
- 173
- Citations by year
- 2026: 13 · 2025: 34 · 2024: 52 · 2023: 29 · 2022: 17 · 2021: 2
- FWCI
- 8.84
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Locations
- IEEE Access (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Physical Sciences › Computer Science › Artificial Intelligence › Stochastic Gradient Optimization Techniques
- Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
- Keywords
- federated learning, decentralized machine learning, edge devices, data privacy, medical image analysis, privacy-preserving machine learning, research challenges
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-09.
Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-09.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
