Blockchain-Based Federated Learning: A Systematic Literature Review of Attacks, Privacy Preservation, and System Integration
Abstract
Federated learning (FL) is a framework for distributed machine learning that trains models without sharing local data and protects privacy. FL uses collaborative learning to build privacy-preserving models. However, FL features face problems like private information disclosure, unreliable model parameter uploads, and communication cost. Blockchain, as a decentralized technology, can improve FL performance without a centralized server and address these issues. This study conducted a systematic literature review on blockchain integration in federated learning, analyzing existing FL problems that can be addressed. Through screening, relevant studies were included, and research questions covered security and privacy attacks in traditional federated learning solvable by blockchain, and characteristics of Blockchain-based FL. The latest blockchain-based approaches to federated learning were studied regarding security and privacy, records and rewards, and verification and accountability. Open issues related to Blockchain and FL combination are discussed. Finally, future research directions for robust Blockchain-based FL systems are proposed
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.