Federated Learning for Internet of Things: A Comprehensive Survey

IEEE Communications Surveys & Tutorials, Journal Year: 2021, Volume and Issue: 23(3), P. 1622 - 1658

Published: Jan. 1, 2021

The Internet of Things (IoT) is penetrating many facets our daily life with the proliferation intelligent services and applications empowered by artificial intelligence (AI). Traditionally, AI techniques require centralized data collection processing that may not be feasible in realistic application scenarios due to high scalability modern IoT networks growing privacy concerns. Federated Learning (FL) has emerged as a distributed collaborative approach can enable applications, allowing for training at devices without need sharing. In this article, we provide comprehensive survey emerging FL networks, beginning from an introduction recent advances discussion their integration. Particularly, explore analyze potential enabling wide range services, including sharing, offloading caching, attack detection, localization, mobile crowdsensing, security. We then extensive use various key such smart healthcare, transportation, Unmanned Aerial Vehicles (UAVs), cities, industry. important lessons learned review FL-IoT are also highlighted. complete highlighting current challenges possible directions future research booming area.

Language: Английский

---

Model-Contrastive Federated Learning

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Journal Year: 2021, Volume and Issue: unknown

Published: June 1, 2021

Federated learning enables multiple parties to collaboratively train a machine model without communicating their local data. A key challenge in federated is handle the heterogeneity of data distribution across parties. Although many studies have been proposed address this challenge, we find that they fail achieve high performance image datasets with deep models. In paper, propose MOON: model-contrastive learning. MOON simple and effective framework. The idea utilize similarity between representations correct training individual parties, i.e., conducting contrastive model-level. Our extensive experiments show significantly outperforms other state-of-the-art algorithms on various classification tasks.

Language: Английский

---

Federated Learning for Smart Healthcare: A Survey

ACM Computing Surveys, Journal Year: 2022, Volume and Issue: 55(3), P. 1 - 37

Published: Feb. 3, 2022

Recent advances in communication technologies and the Internet-of-Medical-Things (IOMT) have transformed smart healthcare enabled by artificial intelligence (AI). Traditionally, AI techniques require centralized data collection processing that may be infeasible realistic scenarios due to high scalability of modern networks growing privacy concerns. Federated Learning (FL), as an emerging distributed collaborative paradigm, is particularly attractive for healthcare, coordinating multiple clients (e.g., hospitals) perform training without sharing raw data. Accordingly, we provide a comprehensive survey on use FL healthcare. First, present recent FL, motivations, requirements using The designs are then discussed, ranging from resource-aware secure privacy-aware incentive personalized FL. Subsequently, state-of-the-art review applications key domains, including health management, remote monitoring, medical imaging, COVID-19 detection. Several FL-based projects analyzed, lessons learned also highlighted. Finally, discuss interesting research challenges possible directions future

Language: Английский

---

Dynamic-Fusion-Based Federated Learning for COVID-19 Detection

IEEE Internet of Things Journal, Journal Year: 2021, Volume and Issue: 8(21), P. 15884 - 15891

Published: Feb. 7, 2021

Medical diagnostic image analysis (e.g., CT scan or X-Ray) using machine learning is an efficient and accurate way to detect COVID-19 infections. However, the sharing of images across medical institutions usually prohibited due patients' privacy concerns. This causes issue insufficient data sets for training classification model. Federated emerging privacy-preserving paradigm that produces unbiased global model based on received local updates trained by clients without exchanging clients' data. Nevertheless, default setting federated introduces a huge communication cost transferring can hardly ensure performance when severe heterogeneity exists. To improve efficiency performance, in this article, we propose novel dynamic fusion-based approach First, design architecture systems analyze images. Furthermore, present fusion method dynamically decide participating according their schedule time. In addition, summarize category detection, which be used community analysis. The evaluation results show proposed feasible performs better than terms efficiency, fault tolerance.

Language: Английский

---

Collaborative Federated Learning for Healthcare: Multi-Modal COVID-19 Diagnosis at the Edge

IEEE Open Journal of the Computer Society, Journal Year: 2022, Volume and Issue: 3, P. 172 - 184

Published: Jan. 1, 2022

Despite significant improvements over the last few years, cloud-based healthcare applications continue to suffer from poor adoption due their limitations in meeting stringent security, privacy, and quality of service requirements (such as low latency). The edge computing trend, along with techniques for distributed machine learning such federated learning, has gained popularity a viable solution settings. In this paper, we leverage capabilities medicine by evaluating potential intelligent processing clinical data at edge. We utilized emerging concept clustered (CFL) an automatic COVID-19 diagnosis. evaluate performance proposed framework under different experimental setups on two benchmark datasets. Promising results are obtained both datasets resulting comparable against central baseline where specialized models (i.e., each specific image modality) trained data, 16% 11% overall F1-Scores have been achieved model (using multi-modal data) CFL setup X-ray Ultrasound datasets, respectively. also discussed associated challenges, technologies, available deploying ML privacy delay-sensitive applications.

Language: Английский

---

Federated Deep Learning for Cyber Security in the Internet of Things: Concepts, Applications, and Experimental Analysis

IEEE Access, Journal Year: 2021, Volume and Issue: 9, P. 138509 - 138542

Published: Jan. 1, 2021

In this article, we present a comprehensive study with an experimental analysis of federated deep learning approaches for cyber security in the Internet Things (IoT) applications. Specifically, first provide review learning-based and privacy systems several types IoT applications, including, Industrial IoT, Edge Computing, Drones, Healthcare Things, Vehicles, etc. Second, use blockchain malware/intrusion detection applications is discussed. Then, vulnerabilities systems. Finally, three approaches, namely, Recurrent Neural Network (RNN), Convolutional (CNN), Deep (DNN). For each model, performance centralized under new real traffic datasets, Bot-IoT dataset, MQTTset TON_IoT dataset. The goal article to important information on emerging technologies security. addition, it demonstrates that outperform classic/centralized versions machine (non-federated learning) assuring device data higher accuracy detecting attacks.

Language: Английский

---

Applications of artificial intelligence in battling against covid-19: A literature review

Chaos Solitons & Fractals, Journal Year: 2020, Volume and Issue: 142, P. 110338 - 110338

Published: Oct. 3, 2020

Language: Английский

---

Multi-Disease Prediction Based on Deep Learning: A Survey

Computer Modeling in Engineering & Sciences, Journal Year: 2021, Volume and Issue: 128(2), P. 489 - 522

Published: Jan. 1, 2021

In recent years, the development of artificial intelligence (AI) and gradual beginning AI’s research in medical field have allowed people to see excellent prospects integration AI healthcare. Among them, hot deep learning has shown greater potential applications such as disease prediction drug response prediction. From initial logistic regression model machine model, then today, accuracy been continuously improved, performance all aspects also significantly improved. This article introduces some basic frameworks common diseases, summarizes methods corresponding different diseases. Point out a series problems current prediction, make prospect for future development. It aims clarify effectiveness demonstrates high correlation between The unique feature extraction can still play an important role research.

Language: Английский

---

Federated learning-based AI approaches in smart healthcare: concepts, taxonomies, challenges and open issues

Cluster Computing, Journal Year: 2022, Volume and Issue: 26(4), P. 2271 - 2311

Published: Aug. 17, 2022

Language: Английский

---

Federated Learning of Electronic Health Records to Improve Mortality Prediction in Hospitalized Patients With COVID-19: Machine Learning Approach

JMIR Medical Informatics, Journal Year: 2021, Volume and Issue: 9(1), P. e24207 - e24207

Published: Jan. 5, 2021

Machine learning models require large datasets that may be siloed across different health care institutions. studies focus on COVID-19 have been limited to single-hospital data, which limits model generalizability.

Language: Английский

---