PFLF: Privacy-Preserving Federated Learning Framework for Edge Computing

Zhou, Hao; Yang, Geng; Dai, Hua; Liu, Guoxiu

doi:10.1109/tifs.2022.3174394

Cited by 45 publications

(8 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…FL technologies have witnessed significant advancements over time, leading to practical applications across various domains, including smart healthcare [24], recommendation systems, smart cities [25], banking, edge training [26], and cybersecurity, among others. To cater to the requirements of these diverse applications, efficient central aggregation algorithms are essential to strike an optimal balance between preserving data security and enhancing computational efficiency.…”

Section: Federated Learningmentioning

confidence: 99%

TabFedSL: A Self-Supervised Approach to Labeling Tabular Data in Federated Learning Environments

Wang,

Hu,

Chen

et al. 2024

Mathematics

View full text Add to dashboard Cite

Currently, self-supervised learning has shown effectiveness in solving data labeling issues. Its success mainly depends on having access to large, high-quality datasets with diverse features. It also relies on utilizing the spatial, temporal, and semantic structures present in the data. However, domains such as finance, healthcare, and insurance primarily utilize tabular data formats. This presents challenges for traditional data augmentation methods aimed at improving data quality. Furthermore, the privacy-sensitive nature of these domains complicates the acquisition of the extensive, high-quality datasets necessary for training effective self-supervised models. To tackle these challenges, our proposal introduces a novel framework that combines self-supervised learning with Federated Learning (FL). This approach aims to solve the problem of data-distributed training while ensuring training quality. Our framework improves upon the conventional self-supervised learning data augmentation paradigm by incorporating data labeling through the segmentation of data into subsets. Our framework adds noise by splitting subsets of data and can achieve the same level of centralized learning in a distributed environment. Moreover, we conduct experiments on various public tabular datasets to evaluate our approach. The experimental results showcase the effectiveness and generalizability of our proposed method in scenarios involving unlabeled data and distributed settings.

show abstract

Section: Federated Learningmentioning

confidence: 99%

TabFedSL: A Self-Supervised Approach to Labeling Tabular Data in Federated Learning Environments

Wang,

Hu,

Chen

et al. 2024

Mathematics

View full text Add to dashboard Cite

show abstract

“…Differential privacy is a commonly used privacy-preserving technique in federated learning. Jia et al [27] and Zhou et al [28] used differential privacy to protect the user's data privacy. Based on differential privacy and secure multi-party computation techniques, Mugunthan et al [29] designed a privacypreserving federated learning scheme by applying a combination of these two techniques to federated learning.…”

Section: Related Workmentioning

confidence: 99%

PPBRFL: Privacy-Preserving Byzantine-Robust Federated Learning

Zhou

2024

FCIS

View full text Add to dashboard Cite

Federated learning is a distributed machine learning approach that allows the neural network to be trained without exposing private user data. Despite its advantages, federated learning schemes still face two critical security challenges: user privacy disclosure and Byzantine robustness. The adversary may try to infer the private data from the trained local gradients or compromise the global model update. To tackle the above challenges, we propose PPBRFL, a privacy-preserving Byzantine-robust federated learning scheme. To resist Byzantine attacks, we design a novel Byzantine-robust aggregation method based on cosine similarity, which can guarantee the global model update and improve the model’s classification accuracy. Furthermore, we introduce a reward and penalty mechanism that considers users’ behavior to mitigate the impact of Byzantine users on the global model. To protect user privacy, we utilize symmetric homomorphic encryption to encrypt the users’ trained local models, which requires low computation cost while maintaining model accuracy. We conduct the experimental assessment of the performance of PPBRFL. The experimental results show that PPBRFL maintains model classification accuracy while ensuring privacy preservation and Byzantine robustness compared to traditional federated learning scheme.

show abstract

“…Other recognized issues related to security and privacy in cloud computing include multi-tenancy, confidentiality, and phishing [62]. Meanwhile, the security issues discussed in edge computing are mainly divided into the following four topics: access control [63,64], identity authentication [65,66], data security [67,68], and privacy [69,70].…”

Section: Security Collaborationmentioning

confidence: 99%

A Design and Application of Municipal Service Platform Based on Cloud-Edge Collaboration for Smart Cities

Yang

Lee

et al. 2022

Sensors

View full text Add to dashboard Cite

Information and Communication Technology (ICT) makes cities “smart”, capable of providing advanced municipal services to citizens more efficiently. In the literature, many applications of municipal service platform based on cloud computing and edge computing have been proposed, but the reference model and application instance based on cloud-edge collaboration specially for municipal service platform is rarely studied. In this context, this paper first develops a reference model, including resource collaboration, application collaboration, service collaboration, and security collaboration, and discusses the main contents and challenges of each part. Then, aiming at the problem of computing and communication resources allocation in the cloud-edge collaboration, a game-theory-based dynamic resource allocation model is introduced. Finally, an e-government self-service system based on the cloud-edge collaboration is designed and implemented. The cloud side is a cloud computing server, and the edge side are the self-service terminals integrating various edge computing devices with Artificial Intelligence (AI) embedded. The experimental results show that the designed system combines the advantages of cloud computing and edge computing, and provides a better user experience with lower processing latency, larger bandwidth, and more concurrent tasks. Meanwhile, the findings show that the evolutionary equilibrium and the Nash equilibrium are the optimal solutions, respectively.

show abstract

PFLF: Privacy-Preserving Federated Learning Framework for Edge Computing

Cited by 45 publications

References 35 publications

TabFedSL: A Self-Supervised Approach to Labeling Tabular Data in Federated Learning Environments

TabFedSL: A Self-Supervised Approach to Labeling Tabular Data in Federated Learning Environments

PPBRFL: Privacy-Preserving Byzantine-Robust Federated Learning

A Design and Application of Municipal Service Platform Based on Cloud-Edge Collaboration for Smart Cities

Contact Info

Product

Resources

About