mlsanityData to 14 Oct 2026

Magnitude-based Neuron Pruning for Backdoor Defens

arXiv 2405.17750 ↗First posted 28 May 2024cs.LG cs.AI cs.CRcited by 1

Not on the map. Why?

Abstract (arXiv)

Deep Neural Networks (DNNs) are known to be vulnerable to backdoor attacks, posing concerning threats to their reliable deployment. Recent research reveals that backdoors can be erased from infected DNNs by pruning a specific group of neurons, while how to effectively identify and remove these backdoor-associated neurons remains an open challenge. In this paper, we investigate the correlation between backdoor behavior and neuron magnitude, and find that backdoor neurons deviate from the magnitude-saliency correlation of the model. The deviation inspires us to propose a Magnitude-based Neuron Pruning (MNP) method to detect and prune backdoor neurons. Specifically, MNP uses three magnitude-guided objective functions to manipulate the magnitude-saliency correlation of backdoor neurons, thus achieving the purpose of exposing backdoor behavior, eliminating backdoor neurons and preserving clean neurons, respectively. Experiments show our pruning strategy achieves state-of-the-art backdoor defense performance against a variety of backdoor attacks with a limited amount of clean data, demonstrating the crucial role of magnitude for guiding backdoor defenses.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 27 of its 39 references that are in the map. First posted on arXiv 28 May 2024.

Builds on

27 of 39 references.

References · 27

  1. 2016Pruning Filters for Efficient ConvNetscited by 1,143
  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. 2018Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networkscited by 580
  5. 2019Importance Estimation for Neural Network Pruningcited by 541
  6. 2018Poison Frogs! Targeted Clean-Label Poisoning Attacks on Neural Networkscited by 519
  7. 2018Spectral Signatures in Backdoor Attackscited by 484
  8. 2019STRIP: A Defence Against Trojan Attacks on Deep Neural Networkscited by 459
  9. 2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 456
  10. 2021Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networkscited by 324
  11. 2018Soft Filter Pruning for Accelerating Deep Convolutional Neural Networkscited by 315
  12. 2020Input-Aware Dynamic Backdoor Attackcited by 313
15 more
  1. 2021Anti-Backdoor Learning: Training Clean Models on Poisoned Datacited by 263
  2. 2019A new Backdoor Attack in CNNs by training set corruption without label poisoningcited by 242
  3. 2021Adversarial Neuron Pruning Purifies Backdoored Deep Modelscited by 235
  4. 2021Adversarial Unlearning of Backdoors via Implicit Hypergradientcited by 166
  5. 2019Universal Litmus Patterns: Revealing Backdoor Attacks in CNNscited by 146
  6. 2020Deep Feature Space Trojan Attack of Neural Networks by Controlled Detoxificationcited by 95
  7. 2022Data-free Backdoor Removal based on Channel Lipschitznesscited by 79
  8. 2023Enhancing Fine-Tuning Based Backdoor Defense with Sharpness-Aware Minimizationcited by 64
  9. 2023Reconstructive Neuron Pruning for Backdoor Defensecited by 59
  10. 2020Can Adversarial Weight Perturbations Inject Neural Backdoors?cited by 56
  11. 2021Trigger Hunting with a Topological Prior for Trojan Detectioncited by 39
  12. 2022Rethinking the Reverse-engineering of Trojan Triggerscited by 39
  13. 2022Training with More Confidence: Mitigating Injected and Natural Backdoors During Trainingcited by 38
  14. 2021Towards Practical Deployment-Stage Backdoor Attack on Deep Neural Networkscited by 34
  15. 2022One-shot Neural Backdoor Erasing via Adversarial Weight Maskingcited by 28

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025Gradient Norm-based Fine-Tuning for Backdoor Defense in Automatic Speech Recognitioncited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2024Rethinking Pruning for Backdoor Mitigation: An Optimization Perspective
  2. 2022Defense against Backdoor Attacks via Identifying and Purifying Bad Neurons
  3. 2024Fusing Pruned and Backdoored Models: Optimal Transport-based Data-free Backdoor Mitigation
  4. 2025Backdoor Mitigation via Invertible Pruning Masks
  5. 2024Unveiling and Mitigating Backdoor Vulnerabilities based on Unlearning Weight Changes and Backdoor Activeness
  6. 2021Few-shot Backdoor Defense Using Shapley Estimation
  7. 2021Red Alarm for Pre-trained Models: Universal Vulnerability to Neuron-Level Backdoor Attacks
  8. 2024BAN: Detecting Backdoors Activated by Adversarial Neuron Noise