Data-Centric Client Selection for Federated Learning Over Distributed Edge Networks

Saha, Rituparna; Misra, Sudip; Chakraborty, Aishwariya; Chatterjee, Chandranath; Deb, Pallav Kumar

doi:10.1109/tpds.2022.3217271

Cited by 26 publications

(9 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In greedy or dynamic methods, resource constraint issues [ 87 , 88 ] contain bandwidth allocation issues [ 5 , 40 ], communication cost issues [ 33 , 85 ], limited computational resources issues [ 8 , 42 , 85 ], and the energy consumption of selected clients [ 26 , 42 ], which can lead to low accuracy and high convergence time and latency. The authors in [ 40 ] proposed a novel perspective to resource allocation in WFLNs, realizing that learning rounds are temporally interdependent and have varying significance toward the final learning outcome.…”

Section: Pros and Cons Of Different Cs Methodsmentioning

confidence: 99%

“…The authors studied diverse scheduling models to select an appropriate participant client in the learning process at each round. In contrast, the authors in [ 42 ] prefer to choose high-data quality clients, ensuring system efficiency and prioritizing the clients who have suitable data rates rather than those with poor calculation and transmission capacities. So, it optimizes on-device data quality across clients while reducing delay, energy consumption, and packet size.…”

Section: Pros and Cons Of Different Cs Methodsmentioning

confidence: 99%

“…In random selection, which is the conventional and basic form of CS, resource constraint issues [ 87 , 88 ] contain bandwidth allocation issues [ 5 ], limited computational resources issues [ 42 ], and the energy consumption of selected clients [ 26 , 42 ], which can lead to low accuracy and high convergence time and latency. In [ 26 ], the authors offered a novel strategy to choose fewer clients in earlier global iterations and more clients in later global iterations in the same period of training.…”

Section: Pros and Cons Of Different Cs Methodsmentioning

confidence: 99%

“…The increasing number and type of network services and the proliferation of mobile edge have prompted the deployment of IoT [ 2 ] devices with advanced sensors, computing, and communication capabilities for crowd-sensing tasks [ 42 ]. The advent of AI has led to significant developments in numerous modern applications, such as air quality, weather monitoring, and video surveillance [ 1 , 2 ].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Comprehensive Overview of IoT-Based Federated Learning: Focusing on Client Selection Methods

Khajehali,

Yan,

Chow

et al. 2023

Sensors

View full text Add to dashboard Cite

The integration of the Internet of Things (IoT) with machine learning (ML) is revolutionizing how services and applications impact our daily lives. In traditional ML methods, data are collected and processed centrally. However, modern IoT networks face challenges in implementing this approach due to their vast amount of data and privacy concerns. To overcome these issues, federated learning (FL) has emerged as a solution. FL allows ML methods to achieve collaborative training by transferring model parameters instead of client data. One of the significant challenges of federated learning is that IoT devices as clients usually have different computation and communication capacities in a dynamic environment. At the same time, their network availability is unstable, and their data quality varies. To achieve high-quality federated learning and handle these challenges, designing the proper client selection process and methods are essential, which involves selecting suitable clients from the candidates. This study presents a comprehensive systematic literature review (SLR) that focuses on the challenges of client selection (CS) in the context of federated learning (FL). The objective of this SLR is to facilitate future research and development of CS methods in FL. Additionally, a detailed and in-depth overview of the CS process is provided, encompassing its abstract implementation and essential characteristics. This comprehensive presentation enables the application of CS in diverse domains. Furthermore, various CS methods are thoroughly categorized and explained based on their key characteristics and their ability to address specific challenges. This categorization offers valuable insights into the current state of the literature while also providing a roadmap for prospective investigations in this area of research.

show abstract

Section: Pros and Cons Of Different Cs Methodsmentioning

confidence: 99%

Section: Pros and Cons Of Different Cs Methodsmentioning

confidence: 99%

Section: Pros and Cons Of Different Cs Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Comprehensive Overview of IoT-Based Federated Learning: Focusing on Client Selection Methods

Khajehali,

Yan,

Chow

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…The results of REFL revealed that the quality of the model was improved with 2

\times

less resource usage as compared to the state‐of‐the‐art. A client selection scheme (DICE) for distributed edge network was proposed in Saha et al (2023). DICE prioritized clients with high‐quality data throughout the selection step in addition to their computational and networking capabilities in order to improve FL accuracy.…”

Section: Client Selection In Flmentioning

confidence: 99%

A review on client selection models in federated learning

Panigrahi,

Bharti,

Sharma

2023

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Federated learning (FL) is a decentralized machine learning (ML) technique that enables multiple clients to collaboratively train a common ML model without them having to share their raw data with each other. A typical FL process involves (1) FL client(s) selection, (2) global model distribution, (3) local training, and (4) aggregation. As such FL clients are heterogeneous edge devices (i.e., mobile phones) that differ in terms of computational resources, training data quality, and distribution. Therefore, FL client(s) selection has a significant influence on the execution of the remaining steps of an FL process. There have been a variety of FL client(s) selection models proposed in the literature, however, their critical review and/or comparative analysis is much less discussed. This paper brings the scattered FL client(s) selection models onto a single platform by first categorizing them into five categories, followed by providing a detailed analysis of the benefits/shortcomings and the applicability of these models for different FL scenarios. Such understanding can help researchers in academia and industry to develop improved FL client(s) selection models to address the requirement challenges and shortcomings of the current models. Finally, future research directions in the area of FL client(s) selection are also discussed.This article is categorized under: Technologies > Machine Learning Technologies > Artificial Intelligence

show abstract

Privacy‐preserving model splitting and quality‐aware device association for federated edge learning

Dong

Shen

et al. 2023

Softw Pract Exp

View full text Add to dashboard Cite

Federated edge learning (FEEL) provides a promising device‐edge collaborative learning paradigm, which enables edge devices to parallel participate in model co‐creation while preserving user privacy, opening countless opportunities to enable edge intelligence. With the growing demand for intelligent services, extensive FEEL deployment is inevitable. Nevertheless, existing FL schemes neglect two unique features (i.e., resource heterogeneity and data heterogeneity) in real‐world edge learning and thus may negatively affect the training efficiency and accuracy. Specifically, (1) heterogeneous and limited device resources cause massive laggards, which bring intolerable training delay; (2) heterogeneous data distribution causes device quality divergence, bringing severe training accuracy degradation. This article proposes a split‐based FEEL framework and an adaptive model splitting and quality‐aware device association scheme (MSDA) to tackle the aforementioned challenges. MSDA contains two levels: at the model splitting level, according to device capability and model structure, an adaptive splitting mechanism is proposed to provide a low‐latency and privacy‐preserving model splitting strategy for each device and guide subsequent device association. At the device association level, each device is simulated as a player with a quality weight in the potential game. Then a quality‐aware decentralized device association mechanism is designed to ensure that more high‐quality devices upload local updates before the deadline with the help of the edge server. Finally, experimental results demonstrate that MSDA yields significant improvements, achieving up to 3.1 training speedup and 39% accuracy improvement compared to state‐of‐the‐art methods.

show abstract

Data-Centric Client Selection for Federated Learning Over Distributed Edge Networks

Cited by 26 publications

References 31 publications

A Comprehensive Overview of IoT-Based Federated Learning: Focusing on Client Selection Methods

A Comprehensive Overview of IoT-Based Federated Learning: Focusing on Client Selection Methods

A review on client selection models in federated learning

Privacy‐preserving model splitting and quality‐aware device association for federated edge learning

Contact Info

Product

Resources

About