mlsanityData to 14 Oct 2026

FedBayes: A Zero-Trust Federated Learning Aggregation to Defend Against Adversarial Attacks

arXiv 2312.04587 ↗First posted 4 Dec 2023cs.CR cs.LGcited by 2

On the map: LLM safety and security → Federated unlearning and poisoning

Abstract (arXiv)

Federated learning has created a decentralized method to train a machine learning model without needing direct access to client data. The main goal of a federated learning architecture is to protect the privacy of each client while still contributing to the training of the global model. However, the main advantage of privacy in federated learning is also the easiest aspect to exploit. Without being able to see the clients' data, it is difficult to determine the quality of the data. By utilizing data poisoning methods, such as backdoor or label-flipping attacks, or by sending manipulated information about their data back to the server, malicious clients are able to corrupt the global model and degrade performance across all clients within a federation. Our novel aggregation method, FedBayes, mitigates the effect of a malicious client by calculating the probabilities of a client's model weights given to the prior model's weights using Bayesian statistics. Our results show that this approach negates the effects of malicious clients and protects the overall federation.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Federated unlearning and poisoning, one of 12 directions in LLM safety and security. Builds on 8 of its 16 references that are in the map. First posted on arXiv 4 Dec 2023.

Builds on

8 of 16 references.

References · 8

  1. 2019Federated Learning: Challenges, Methods, and Future Directionscited by 1,538
  2. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,106
  3. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,055
  4. 2020Adaptive Federated Optimizationcited by 1,046
  5. 2020Flower: A Friendly Federated Learning Research Frameworkcited by 455
  6. 2021Challenges and Approaches for Mitigating Byzantine Attacks in Federated Learningcited by 47
  7. 2021A Bayesian Federated Learning Framework with Online Laplace Approximationcited by 41
  8. 2021Untargeted, Targeted and Universal Adversarial Attacks and Defenses on Time Seriescited by 21

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Trust Aware Federated Learning for Secure Bone Healing Stage Interpretation in e-Healthcited by 0
  2. 2026Agentic Trust Coordination for Federated Learning through Adaptive Thresholding and Autonomous Decision Making in Sustainable and Resilient Industrial Networkscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024No Vandalism: Privacy-Preserving and Byzantine-Robust Federated Learning
  2. 2023FedDefender: Client-Side Attack-Tolerant Federated Learning
  3. 2024FedAA: A Reinforcement Learning Perspective on Adaptive Aggregation for Fair and Robust Federated Learning
  4. 2024Fake or Compromised? Making Sense of Malicious Clients in Federated Learning
  5. 2024Defending against Data Poisoning Attacks in Federated Learning via User Elimination
  6. 2026Mitigating Backdoor Attacks in Federated Learning Using PPA and MiniMax Game Theory
  7. 2025FedGuard: A Diverse-Byzantine-Robust Mechanism for Federated Learning with Major Malicious Clients ↗
  8. 2025SoK: Benchmarking Poisoning Attacks and Defenses in Federated Learning