FedAdapt: Adaptive Offloading for IoT Devices in Federated Learning

Wu, Di; Ullah, Rehmat; Harvey, Paul; Kilpatrick, Peter; Spence, Ivor; Varghese, Blesson

doi:10.1109/jiot.2022.3176469

Cited by 59 publications

(18 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The RL agent uses the average time of each training round as the reward function to minimize the average training time of all devices. The experiment results obtained in [71] indicate that FedAdapt may achieve substantial reduction in average training and can adapt to changes in network bandwidth as well as heterogeneity in IoT devices. On the other hand, FedAdapt introduces extra complexity and overheads caused by the RL agent and the clustering algorithm.…”

Section: B Model Decomposition In Hybrid Sl-fl Frameworkmentioning

confidence: 98%

“…Therefore, static split between the client-and server-side models lacks the flexibility to adapt to a dynamic IoT environment. As an attempt to address this challenge, FedAdapt was proposed in [71] as a hybrid FL-SL framework that is able to adaptively determine which portion of a model to be offloaded to a server based on the computational resources on the client devices and the network bandwidth between clients and the server.…”

Section: B Model Decomposition In Hybrid Sl-fl Frameworkmentioning

confidence: 99%

See 1 more Smart Citation

Combined Federated and Split Learning in Edge Computing for Ubiquitous Intelligence in Internet of Things: State-of-the-Art and Future Directions

Duan

Deng

et al. 2022

Sensors

View full text Add to dashboard Cite

Federated learning (FL) and split learning (SL) are two emerging collaborative learning methods that may greatly facilitate ubiquitous intelligence in the Internet of Things (IoT). Federated learning enables machine learning (ML) models locally trained using private data to be aggregated into a global model. Split learning allows different portions of an ML model to be collaboratively trained on different workers in a learning framework. Federated learning and split learning, each have unique advantages and respective limitations, may complement each other toward ubiquitous intelligence in IoT. Therefore, the combination of federated learning and split learning recently became an active research area attracting extensive interest. In this article, we review the latest developments in federated learning and split learning and present a survey on the state-of-the-art technologies for combining these two learning methods in an edge computing-based IoT environment. We also identify some open problems and discuss possible directions for future research in this area with the hope of arousing the research community’s interest in this emerging field.

show abstract

Section: B Model Decomposition In Hybrid Sl-fl Frameworkmentioning

confidence: 98%

Section: B Model Decomposition In Hybrid Sl-fl Frameworkmentioning

confidence: 99%

Combined Federated and Split Learning in Edge Computing for Ubiquitous Intelligence in Internet of Things: State-of-the-Art and Future Directions

Duan

Deng

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Personalized FL has also been proposed to surmount the challenges arising due to device heterogeneity in a cloud-edge based FL system [25]. FedAdapt [26] adopts the reinforcement learning technique to adjust the amount of computation that is offloaded to the edge server, thereby adapting to varying resource availability and computational capacities of heterogeneous end-user devices. However, above researches rarely consider the effect of different communication protocols for a edge-based FL system since end-user devices and edge nodes always need customized communication protocols to meet their application requirements.…”

Section: A Edge-based Flmentioning

confidence: 99%

“…In an edge-based FL system, the global model is collaboratively trained across a large number of end-user devices with limited network bandwidth and unstable connections [26]. As a result, the communication cost is a key consideration in FL training.…”

Section: Communication Protocols In Flmentioning

confidence: 99%

“…This paper presents a methodology, namely FedComm that benchmarks a variety of lightweight application layer and transport layer protocols within the context of FL. The classic FL implementation suitable for resource constrained devices and can operate in a device-edge environment presented in FedAdapt [10] is chosen for FedComm. Metrics that capture performance aspects, including the accuracy of the model, and communication time in diverse network conditions, namely 3G, 4G and Wi-Fi based networks are considered.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

FedComm: Understanding Communication Protocols for Edge-based Federated Learning

Cleland¹,

Wu²,

Ullah³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Federated learning (FL) trains machine learning (ML) models on devices using locally generated data and exchanges models without transferring raw data to a distant server. This exchange incurs a communication overhead and impacts the performance of FL training. There is limited understanding of how communication protocols specifically contribute to the performance of FL. Such an understanding is essential for selecting the right communication protocol when designing an FL system. This paper presents FedComm a benchmarking methodology to quantify the impact of optimized application layer protocols, namely Message Queue Telemetry Transport (MQTT), Advanced Message Queuing Protocol (AMQP), and Ze-roMQ Message Transport Protocol (ZMTP), and non-optimized application layer protocols, namely as TCP and UDP, on the performance of FL. FedComm measures the overall performance of FL in terms of communication time and accuracy under varying computational and network stress and packet loss rates. Experiments on a lab-based testbed demonstrate that TCP outperforms UDP as a non-optimized application layer protocol with higher accuracy and shorter communication times for 4G and Wi-Fi networks. Optimized application layer protocols such as AMQP, MQTT, and ZMTP outperformed nonoptimized application layer protocols in most network conditions, resulting in a 2.5x reduction in communication time compared to TCP while maintaining accuracy. The experimental results enable us to highlight a number of open research issues for further investigation. FedComm is available for download from https://github.com/qub-blesson/FedComm.

show abstract

Federated Learning for Intelligent IoT Systems: Background, Frameworks, and Optimization Techniques

Ray

2024

Model Optimization Methods for Efficient and Edge AI

View full text Add to dashboard Cite

FedAdapt: Adaptive Offloading for IoT Devices in Federated Learning

Cited by 59 publications

References 43 publications

Combined Federated and Split Learning in Edge Computing for Ubiquitous Intelligence in Internet of Things: State-of-the-Art and Future Directions

Combined Federated and Split Learning in Edge Computing for Ubiquitous Intelligence in Internet of Things: State-of-the-Art and Future Directions

FedComm: Understanding Communication Protocols for Edge-based Federated Learning

Federated Learning for Intelligent IoT Systems: Background, Frameworks, and Optimization Techniques

Contact Info

Product

Resources

About