mlsanityData to 14 Oct 2026

FedNIA: Noise-Induced Activation Analysis for Mitigating Data Poisoning in Federated Learning

arXiv 2502.16396 ↗First posted 23 Feb 2025cs.LG cs.AI cs.CRcited by 0

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

Abstract (arXiv)

Federated learning systems are increasingly threatened by data poisoning attacks, where malicious clients compromise global models by contributing tampered updates. Existing defenses often rely on impractical assumptions, such as access to a central test dataset, or fail to generalize across diverse attack types, particularly those involving multiple malicious clients working collaboratively. To address this, we propose Federated Noise-Induced Activation Analysis (FedNIA), a novel defense framework to identify and exclude adversarial clients without relying on any central test dataset. FedNIA injects random noise inputs to analyze the layerwise activation patterns in client models leveraging an autoencoder that detects abnormal behaviors indicative of data poisoning. FedNIA can defend against diverse attack types, including sample poisoning, label flipping, and backdoors, even in scenarios with multiple attacking nodes. Experimental results on non-iid federated datasets demonstrate its effectiveness and robustness, underscoring its potential as a foundational approach for enhancing the security of federated learning systems.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Federated unlearning and poisoning, one of 12 directions in LLM safety and security. Builds on 32 of its 55 references that are in the map. First posted on arXiv 23 Feb 2025.

Builds on

32 of 55 references.

References · 32

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  3. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,349
  4. 2018How To Backdoor Federated Learningcited by 765
  5. 2018Byzantine-Robust Distributed Learning: Towards Optimal Statistical Ratescited by 711
  6. 2019Local Model Poisoning Attacks to Byzantine-Robust Federated Learningcited by 493
  7. 2020Backdoor Learning: A Surveycited by 426
  8. 2018Analyzing Federated Learning through an Adversarial Lenscited by 351
  9. 2020FLTrust: Byzantine-robust Federated Learning via Trust Bootstrappingcited by 346
  10. 2018The Hidden Vulnerability of Distributed Learning in Byzantiumcited by 329
  11. 2019Robust Aggregation for Federated Learningcited by 317
  12. 2020Attack of the Tails: Yes, You Really Can Backdoor Federated Learningcited by 309
20 more
  1. 2019Can You Really Backdoor Federated Learning?cited by 255
  2. 2019Differentially Private Learning with Adaptive Clippingcited by 220
  3. 2020Auditing Differentially Private Machine Learning: How Private is Private SGD?cited by 218
  4. 2020Privacy and Robustness in Federated Learning: Attacks and Defensescited by 166
  5. 2021Back to the Drawing Board: A Critical Evaluation of Poisoning Attacks on Production Federated Learningcited by 118
  6. 2021CRFL: Certifiably Robust Federated Learning against Backdoor Attackscited by 91
  7. 2020Learning to Detect Malicious Clients for Robust Federated Learningcited by 88
  8. 2020A Blockchain-based Decentralized Federated Learning Framework with Committee Consensuscited by 84
  9. 2020Defending against Backdoors in Federated Learning with Robust Learning Ratecited by 82
  10. 2019Data Poisoning against Differentially-Private Learners: Attacks and Defensescited by 78
  11. 2020Local and Central Differential Privacy for Robustness and Privacy in Federated Learningcited by 78
  12. 2022MPAF: Model Poisoning Attacks to Federated Learning based on Fake Clientscited by 68
  13. 2019Abnormal Client Behavior Detection in Federated Learningcited by 40
  14. 2021FL-WBC: Enhancing Robustness against Model Poisoning Attacks in Federated Learning from a Client Perspectivecited by 36
  15. 2024Decentralized Federated Learning: A Survey on Security and Privacycited by 25
  16. 2019PIRATE: A Blockchain-based Secure Framework of Distributed Machine Learning in 5G Networkscited by 18
  17. 2023Data and Model Poisoning Backdoor Attacks on Wireless Federated Learning, and the Defense Mechanisms: A Comprehensive Surveycited by 18
  18. 2022BEAS: Blockchain Enabled Asynchronous & Secure Federated Machine Learningcited by 7
  19. 2021Double-Adversarial Activation Anomaly Detection: Adversarial Autoencoders are Anomaly Generatorscited by 3
  20. 2024Enhancing Security and Privacy in Federated Learning using Low-Dimensional Update Representation and Proximity-Based Defensecited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Federated Learning Under Attack: Exposing Vulnerabilities through Data Poisoning Attacks in Computer Networks
  2. 2024Precision Guided Approach to Mitigate Data Poisoning Attacks in Federated Learning
  3. 2025SoK: Benchmarking Poisoning Attacks and Defenses in Federated Learning
  4. 2023FreqFed: A Frequency Analysis-Based Approach for Mitigating Poisoning Attacks in Federated Learning
  5. 2022FedCC: Robust Federated Learning against Model Poisoning Attacks
  6. 2026FedSurrogate: Backdoor Defense in Federated Learning via Layer Criticality and Surrogate Replacement
  7. 2024Data Poisoning and Leakage Analysis in Federated Learning
  8. 2024Mitigating Malicious Attacks in Federated Learning via Confidence-aware Defense