Defending against Backdoors in Federated Learning with Robust Learning Rate
Not on the map. Why?
Abstract (arXiv)
Federated learning (FL) allows a set of agents to collaboratively train a model without sharing their potentially sensitive data. This makes FL suitable for privacy-preserving applications. At the same time, FL is susceptible to adversarial attacks due to decentralized and unvetted data. One important line of attacks against FL is the backdoor attacks. In a backdoor attack, an adversary tries to embed a backdoor functionality to the model during training that can later be activated to cause a desired misclassification. To prevent backdoor attacks, we propose a lightweight defense that requires minimal change to the FL protocol. At a high level, our defense is based on carefully adjusting the aggregation server's learning rate, per dimension and per round, based on the sign information of agents' updates. We first conjecture the necessary steps to carry a successful backdoor attack in FL setting, and then, explicitly formulate the defense based on our conjecture. Through experiments, we provide empirical evidence that supports our conjecture, and we test our defense against backdoor attacks under different settings. We observe that either backdoor is completely eliminated, or its accuracy is significantly reduced. Overall, our experiments suggest that our defense significantly outperforms some of the recently proposed defenses in the literature. We achieve this by having minimal influence over the accuracy of the trained models. In addition, we also provide convergence rate analysis for our proposed scheme.
Key facts. Cited by 82 papers in the map (data through 14 Oct 2026). First posted on arXiv 7 Jul 2020.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 2; 2022: 10; 2023: 27; 2024: 14; 2025: 22; 2026: 7.
82 papers.
Sort:
By area:
- 2023Heterogeneous Federated Learning: State-of-the-art and Research Challengescited by 161
- 2024Vaccine: Perturbation-aware Alignment for Large Language Models against Harmful Fine-tuning Attackcited by 115
- 2023Federated Learning for Generalization, Robustness, Fairness: A Survey and Benchmarkcited by 99
- 2024Lisa: Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning Attackcited by 92
- 2022Survey on Federated Learning Threats: concepts, taxonomy on attacks and defences, experimental study and challengescited by 84
- 2023Recent Advances on Federated Learning: A Systematic Surveycited by 41
- 2023Controlling the Extraction of Memorized Data from Large Language Models via Prompt-Tuningcited by 40
- 2022FLIP: A Provable Defense Framework for Backdoor Mitigation in Federated Learningcited by 37
- 2021Emerging Trends in Federated Learning: From Model Fusion to Federated X Learningcited by 35
- 2023Backdoor Attacks and Defenses in Federated Learning: Survey, Challenges and Future Research Directionscited by 33
- 2023On the Vulnerability of Backdoor Defenses for Federated Learningcited by 27
- 2023Multi-metrics adaptively identifies backdoors in Federated learningcited by 23
38 more
- 2023FedSecurity: Benchmarking Attacks and Defenses in Federated Learning and Federated LLMscited by 21
- 2023Backdoor Federated Learning by Poisoning Backdoor-Critical Layerscited by 18
- 2023Data and Model Poisoning Backdoor Attacks on Wireless Federated Learning, and the Defense Mechanisms: A Comprehensive Surveycited by 18
- 2024Threats and Defenses in Federated Learning Life Cycle: A Comprehensive Survey and Challengescited by 15
- 2023Defenses in Adversarial Machine Learning: A Surveycited by 14
- 2023Get Rid Of Your Trail: Remotely Erasing Backdoors in Federated Learningcited by 13
- 2025Detecting Backdoor Attacks in Federated Learning via Direction Alignment Inspectioncited by 13
- 2022Backdoor Attack and Defense in Federated Generative Adversarial Network-based Medical Image Synthesiscited by 12
- 2023G2uardFL: Safeguarding Federated Learning Against Backdoor Attacks through Attributed Client Graph Clusteringcited by 9
- 2023A Survey on Vulnerability of Federated Learning: A Learning Algorithm Perspectivecited by 9
- 2024Meta Stackelberg Game: Robust Federated Learning against Adaptive and Mixed Poisoning Attackscited by 9
- 2022Backdoor Attack is a Devil in Federated GAN-based Medical Image Synthesiscited by 7
- 2022Long-Short History of Gradients is All You Need: Detecting Malicious and Unreliable Clients in Federated Learningcited by 7
- 2023FedDebug: Systematic Debugging for Federated Learning Applicationscited by 6
- 2023Resisting Backdoor Attacks in Federated Learning via Bidirectional Elections and Individual Perspectivecited by 6
- 2025Byzantine-Robust Federated Learning over Ring-All-Reduce Distributed Computingcited by 6
- 2023FilterFL: Knowledge Filtering-based Data-Free Backdoor Defense for Federated Learningcited by 5
- 2024ACE: A Model Poisoning Attack on Contribution Evaluation Methods in Federated Learningcited by 5
- 2022Federated Zero-Shot Learning for Visual Recognitioncited by 4
- 2023FedGrad: Mitigating Backdoor Attacks in Federated Learning Through Local Ultimate Gradients Inspectioncited by 4
- 2023Bkd-FedGNN: A Benchmark for Classification Backdoor Attacks on Federated Graph Neural Networkcited by 4
- 2023FedDefender: Backdoor Attack Defense in Federated Learningcited by 4
- 2024Identify Backdoored Model in Federated Learning via Individual Unlearningcited by 4
- 2025Toward Malicious Clients Detection in Federated Learningcited by 4
- 2024Federated Learning for Smart Grid: A Survey on Applications and Potential Vulnerabilitiescited by 3
- 2022FedCos: A Scene-adaptive Federated Optimization Enhancement for Performance Improvementcited by 2
- 2022Invariant Aggregator for Defending against Federated Backdoor Attackscited by 2
- 2022Resilience of Wireless Ad Hoc Federated Learning against Model Poisoning Attackscited by 2
- 2023Keep It Simple: Fault Tolerance Evaluation of Federated Learning with Unreliable Clientscited by 2
- 2023Kick Bad Guys Out! Conditionally Activated Anomaly Detection in Federated Learning with Zero-Knowledge Proof Verificationcited by 2
- 2025Robust Knowledge Distillation in Federated Learning: Counteracting Backdoor Attackscited by 2
- 2025Multi-Target Federated Backdoor Attack Based on Feature Aggregationcited by 2
- 2021The Impact of Data Distribution on Fairness and Robustness in Federated Learningcited by 1
- 2023Adversarial Robustness Unhardening via Backdoor Attacks in Federated Learningcited by 1
- 2024FLARE: A New Federated Learning Framework with Adjustable Learning Rates over Resource-Constrained Wireless Networkscited by 1
- 2024Concealing Backdoor Model Updates in Federated Learning by Trigger-Optimized Data Poisoningcited by 1
- 2024Venomancer: Towards Imperceptible and Target-on-Demand Backdoor Attacks in Federated Learningcited by 1
- 2024TPFL: A Trustworthy Personalized Federated Learning Framework via Subjective Logiccited by 1
50 most cited of 82 All 82 (JSON)
- 2026STAR-FL: Secure Federated Learning with Spatial-Temporal Analysis and Robust Aggregation14 Aug 2026
- 2026Defending Against Backdoor Attacks via Alignment Checking in Model-Contrastive Federated Learning29 Jul 2026
- 2026FedLSG: LLM-Enhanced Semantic Calibration for Federated Graph Backdoor Defense22 Jul 2026
- 2026PoiCGAN: A Targeted Poisoning Based on Feature-Label Joint Perturbation in Federated Learning24 Mar 2026
- 2026TASER: Task-Aware Spectral Energy Refine for Backdoor Suppression in UAV Swarms Decentralized Federated Learning10 Mar 2026
- 2026Exploiting Layer-Specific Vulnerabilities to Backdoor Attack in Federated Learning16 Feb 2026
- 2026FAROS: Robust Federated Learning with Adaptive Scaling against Backdoor Attacks5 Jan 2026
- 2025Unveiling Hidden Threats: Using Fractal Triggers to Boost Stealthiness of Distributed Backdoor Attacks in Federated Learning12 Nov 2025
- 2025Power to the Clients: Federated Learning in a Dictatorship Setting25 Oct 2025
- 2025POLAR: Policy-based Layerwise Reinforcement Learning Method for Stealthy Backdoor Attacks in Federated Learning21 Oct 2025
- 2025On the Out-of-Distribution Backdoor Attack for Federated Learning16 Sep 2025
- 2025Hammer and Anvil: Toward a Theory of Backdoors in Federated Learning9 Sep 2025
38 more
- 2025FLAT: Revealing Hidden Latent-Conditioned Backdoor Failures in Federated Learning6 Aug 2025
- 2025Coward: Collision-based OOD Watermarking for Practical Proactive Federated Backdoor Detection4 Aug 2025
- 2025FedBAP: Backdoor Defense via Benign Adversarial Perturbation in Federated Learning26 Jul 2025
- 2025BackFed: An Efficient & Standardized Benchmark Suite for Backdoor Attacks in Federated Learning7 Jul 2025
- 2025Private Aggregation for Byzantine-Resilient Heterogeneous Federated Learning11 Jun 2025
- 2025ATR-Bench: A Federated Learning Benchmark for Adaptation, Trust, and Reasoning22 May 2025
- 2025FedGraM: Defending Against Untargeted Attacks in Federated Learning via Embedding Gram Matrix20 May 2025
- 2025FL-PLAS: Federated Learning with Partial Layer Aggregation for Backdoor Defense Against High-Ratio Malicious Clients17 May 2025
- 2025Defending the Edge: Representative-Attention Defense against Backdoor Attacks in Federated Learning15 May 2025
- 2025Toward Malicious Clients Detection in Federated Learning14 May 2025
- 2025A Client-level Assessment of Collaborative Backdoor Poisoning in Non-IID Federated Learning17 Apr 2025
- 2025Detecting Backdoor Attacks in Federated Learning via Direction Alignment Inspection11 Mar 2025
- 2025Multi-Target Federated Backdoor Attack Based on Feature Aggregation23 Feb 2025
- 2025FedNIA: Noise-Induced Activation Analysis for Mitigating Data Poisoning in Federated Learning23 Feb 2025
- 2025Decoding FL Defenses: Systemization, Pitfalls, and Remedies3 Feb 2025
- 2025Robust Knowledge Distillation in Federated Learning: Counteracting Backdoor Attacks1 Feb 2025
- 2025Byzantine-Robust Federated Learning over Ring-All-Reduce Distributed Computing29 Jan 2025
- 2024LOGSAFE: Logic-Guided Verification for Trustworthy Federated Time-Series Learning5 Nov 2024
- 2024Identify Backdoored Model in Federated Learning via Individual Unlearning1 Nov 2024
- 2024Meta Stackelberg Game: Robust Federated Learning against Adaptive and Mixed Poisoning Attacks22 Oct 2024
- 2024TPFL: A Trustworthy Personalized Federated Learning Framework via Subjective Logic16 Oct 2024
- 2024RAB2-DEF: Dynamic and explainable defense against adversarial attacks in Federated Learning to fair poor clients10 Oct 2024
- 2024Federated Learning for Smart Grid: A Survey on Applications and Potential Vulnerabilities16 Sep 2024
- 2024FLAIN: Mitigating Backdoor Attacks in Federated Learning via Flipping Weight Updates of Low-Activation Input Neurons16 Aug 2024
- 2024Threats and Defenses in Federated Learning Life Cycle: A Comprehensive Survey and Challenges9 Jul 2024
- 2024Venomancer: Towards Imperceptible and Target-on-Demand Backdoor Attacks in Federated Learning3 Jul 2024
- 2024ACE: A Model Poisoning Attack on Contribution Evaluation Methods in Federated Learning31 May 2024
- 2024Lisa: Lazy Safety Alignment for Large Language Models against Harmful Fine-tuning Attack28 May 2024
- 2024Concealing Backdoor Model Updates in Federated Learning by Trigger-Optimized Data Poisoning10 May 2024
- 2024FLARE: A New Federated Learning Framework with Adjustable Learning Rates over Resource-Constrained Wireless Networks23 Apr 2024
- 2024Vaccine: Perturbation-aware Alignment for Large Language Models against Harmful Fine-tuning Attack2 Feb 2024
- 2023Data and Model Poisoning Backdoor Attacks on Wireless Federated Learning, and the Defense Mechanisms: A Comprehensive Survey14 Dec 2023
- 2023Defenses in Adversarial Machine Learning: A Survey13 Dec 2023
- 2023A Survey on Vulnerability of Federated Learning: A Learning Algorithm Perspective27 Nov 2023
- 2023Federated Learning for Generalization, Robustness, Fairness: A Survey and Benchmark12 Nov 2023
- 2023Adversarial Robustness Unhardening via Backdoor Attacks in Federated Learning17 Oct 2023
- 2023Kick Bad Guys Out! Conditionally Activated Anomaly Detection in Federated Learning with Zero-Knowledge Proof Verification6 Oct 2023
- 2023Resisting Backdoor Attacks in Federated Learning via Bidirectional Elections and Individual Perspective28 Sep 2023
50 newest of 82 All 82 (JSON)
- 2022Backdoor Attack and Defense in Federated Generative Adversarial Network-based Medical Image Synthesiscited by 12
- 2022Federated Zero-Shot Learning for Visual Recognitioncited by 4
- 2023Kick Bad Guys Out! Conditionally Activated Anomaly Detection in Federated Learning with Zero-Knowledge Proof Verificationcited by 2
- 2025FedNIA: Noise-Induced Activation Analysis for Mitigating Data Poisoning in Federated Learningcited by 0
- 2026FedLSG: LLM-Enhanced Semantic Calibration for Federated Graph Backdoor Defensecited by 0
- 2026STAR-FL: Secure Federated Learning with Spatial-Temporal Analysis and Robust Aggregationcited by 0
- 2026STAR-FL: Secure Federated Learning with Spatial-Temporal Analysis and Robust Aggregation14 Aug 2026
- 2026FedLSG: LLM-Enhanced Semantic Calibration for Federated Graph Backdoor Defense22 Jul 2026
- 2025FedNIA: Noise-Induced Activation Analysis for Mitigating Data Poisoning in Federated Learning23 Feb 2025
- 2023Kick Bad Guys Out! Conditionally Activated Anomaly Detection in Federated Learning with Zero-Knowledge Proof Verification6 Oct 2023
- 2022Backdoor Attack and Defense in Federated Generative Adversarial Network-based Medical Image Synthesis19 Oct 2022
- 2022Federated Zero-Shot Learning for Visual Recognition5 Sep 2022
- 2023Heterogeneous Federated Learning: State-of-the-art and Research Challengescited by 161
- 2023Federated Learning for Generalization, Robustness, Fairness: A Survey and Benchmarkcited by 99
- 2022Survey on Federated Learning Threats: concepts, taxonomy on attacks and defences, experimental study and challengescited by 84
- 2023Recent Advances on Federated Learning: A Systematic Surveycited by 41
- 2023Controlling the Extraction of Memorized Data from Large Language Models via Prompt-Tuningcited by 40
- 2022FLIP: A Provable Defense Framework for Backdoor Mitigation in Federated Learningcited by 37
- 2021Emerging Trends in Federated Learning: From Model Fusion to Federated X Learningcited by 35
- 2023Backdoor Attacks and Defenses in Federated Learning: Survey, Challenges and Future Research Directionscited by 33
- 2023On the Vulnerability of Backdoor Defenses for Federated Learningcited by 27
- 2023Multi-metrics adaptively identifies backdoors in Federated learningcited by 23
- 2023FedSecurity: Benchmarking Attacks and Defenses in Federated Learning and Federated LLMscited by 21
- 2023Backdoor Federated Learning by Poisoning Backdoor-Critical Layerscited by 18
12 most cited of 71 All 71 (JSON)
- 2026Defending Against Backdoor Attacks via Alignment Checking in Model-Contrastive Federated Learning29 Jul 2026
- 2026PoiCGAN: A Targeted Poisoning Based on Feature-Label Joint Perturbation in Federated Learning24 Mar 2026
- 2026TASER: Task-Aware Spectral Energy Refine for Backdoor Suppression in UAV Swarms Decentralized Federated Learning10 Mar 2026
- 2026Exploiting Layer-Specific Vulnerabilities to Backdoor Attack in Federated Learning16 Feb 2026
- 2026FAROS: Robust Federated Learning with Adaptive Scaling against Backdoor Attacks5 Jan 2026
- 2025Unveiling Hidden Threats: Using Fractal Triggers to Boost Stealthiness of Distributed Backdoor Attacks in Federated Learning12 Nov 2025
- 2025Power to the Clients: Federated Learning in a Dictatorship Setting25 Oct 2025
- 2025On the Out-of-Distribution Backdoor Attack for Federated Learning16 Sep 2025
- 2025Hammer and Anvil: Toward a Theory of Backdoors in Federated Learning9 Sep 2025
- 2025FLAT: Revealing Hidden Latent-Conditioned Backdoor Failures in Federated Learning6 Aug 2025
- 2025Coward: Collision-based OOD Watermarking for Practical Proactive Federated Backdoor Detection4 Aug 2025
- 2025FedBAP: Backdoor Defense via Benign Adversarial Perturbation in Federated Learning26 Jul 2025
12 newest of 71 All 71 (JSON)