Exploring Machine Learning Models for Federated Learning - A Review of Approaches, Performance, and Limitations

Aug 3, 2026

Upload by Aurora

Federated learning is a privacy-preserving distributed learning framework that supports collaborative model training without exposing sensitive individual data. This paper conducts a systematic review of recent privacy-preserving machine learning literature using PRISMA guidelines, focusing on supervised and unsupervised algorithms, ensemble and meta-heuristic methods, blockchain technologies, and reinforcement learning within federated learning. It also summarizes federated learning components and real-world applications, offering a machine-learning perspective for researchers and practitioners, and highlighting open problems and future research directions.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages30

Reading Time1 h 16 min

AccessFree to download

Introduction

Federated Learning and Privacy-Preserving Concepts

Federated Learning Framework and Architecture

Applications and Open Problems

Open Problems and Future Research Directions

arXiv :2311 . 10832v1 [ cs .LG] 17 Nov 2023 Exploring Machine Learning Models for Federated Learning: A Review of Approaches, Performance, and Limitations Elaheh Jafarigol 1*, Theodore B. Trafalis2 , Talayeh Razzaghi2 , Mona Zamankhani3 1* Data Science and Analytics Institute, University of Oklahoma, 202 W. Boyd St., Room 409, Norman, 73019, Ok, USA. 2 School of Industrial and Systems Engineering, University of Oklahoma, 202 W Boyd St., Room 124, Norman, 73019, OK, USA. 3 Department of Industrial Engineering, Isfahan University of Technology, Isfahan, Iran. *Corresponding author(s). E-mail(s): [elaheh.jafarigol@ou.edu](elaheh.jafarigol@ou.edu) ; Abstract In the growing world of artificial intelligence, federated learning is a distributed learning framework enhanced to preserve the privacy of individuals’ data. Federated learning lays the groundwork for collaborative research in areas where the data is sensitive. Federated learning has several implications for real-world problems. In times of crisis, when real-time decision-making is critical, federated learning allows multiple entities to work collectively without sharing sensitive data. This distributed approach enables us to leverage information from multiple sources and gain more diverse insights. This paper is a systematic review of the literature on privacy-preserving machine learning in the last few years based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Specifically, we have presented an extensive review of supervised/unsupervised machine learning algorithms, ensemble methods, meta-heuristic approaches, blockchain technology, and reinforcement learning used in the framework of federated learning, in addition to an overview of federated learning applications. This paper reviews the literature on the components of federated learning and its applications in the last few years. The main purpose of this work is to provide researchers and practitioners with a comprehensive overview of federated learning from the machine learning point of view. A discussion of some open problems and future research directions in federated learning is also provided. Keywords: Federated Learning, Privacy-preserving Machine Learning, Distributed Learning, Supervised/Unsupervised Learning, Artificial Intelligence 1 Introduction Privacy is the individuals’ right to control their personal information. With the advancesin data-driven technologies, ensuring privacy protection has become more challenging. Privacy and federated learning are intertwined concepts in machine learning. Privacy and federated learning are intertwined concepts in machine learning. Federated learning offers a promising solution by enabling collaborative model training without exposing sensitive data. This decentralized approach preserves privacy by design, allowing organizations and individuals to collaborate while minimizing the risk of data breaches or unauthorized access. With privacy at its core, federated learning has become a possible solution for scenarios where the data is sensitive and individual privacy is a concern. An algorithm is considered private if the outcome of the analysis is the same, whether any arbitrary individual data is part of the dataset or not [1] . This definition is the basis for privacy protection mechanisms in federated learning. Federated learning is a broad term that includes different aspects of data collection, storage, analysis, and communication in a decentralized information system where data centers can not disclose data for learning purposes. The term federated learning was introduced in the paper published based on the results of a research project carried out at Google in 2016 for text input prediction on mobile devices [2] . The authors designed a collaborative environment fora group of devices referred to as clients, coordinated by a central server, also known asa service provider. They also cond

FAQs

What problem does federated learning address regarding data privacy?

Federated learning enables collaborative model training while keeping sensitive data on local entities, reducing the risk of data breaches or unauthorized access. Privacy is preserved by design through decentralized learning and limited private communication with a central server.

How is the review conducted in this paper?

The paper is a systematic review of recent literature on privacy-preserving machine learning, following the PRISMA guidelines. It synthesizes approaches, performance considerations, and limitations reported in the prior work.

Which machine learning approach categories are covered for federated learning?

The review covers supervised and unsupervised learning algorithms, ensemble methods, meta-heuristic approaches, blockchain technology, and reinforcement learning within the federated learning framework. It also provides an overview of federated learning applications.

Aug 3, 2026

Upload by Aurora