On the security and privacy of federated learning: A survey with attacks, defenses, frameworks, applications, and future directions
Sapienza University of Rome · Universidad de Murcia
Abstract
• Unified security–privacy taxonomy across FL attacks, defenses, and phases. • Attack–defense map: poisoning, backdoor, GAN inference, Sybil; GAN-defense gaps. • Assessment of 13 FL frameworks; notes limits in non-horizontal FL and metrics. • Survey of FL uses across 12 sectors; domain-specific risks and defense efficacy. • Future agenda: adaptive defenses, hybrid crypto, fairness, scalable verifiable aggregation. Federated Learning (FL) is an emerging distributed machine learning paradigm enabling multiple clients to train a global model collaboratively without sharing their raw data. While FL enhances data privacy by design, it remains vulnerable to various security and privacy threats. This survey provides…
Citation impact
- FWCI
- 98.39
- Percentile
- 99%
- References
- 78
Authors
6Topics & keywords
- Federated learning
- Verifiable secret sharing
- Scalability
- Safeguarding
- Robustness (evolution)
- Open research
- Data sharing
- Differential privacy