mlsanityData to 7 Oct 2026

TOP: Backdoor Detection in Neural Networks via Transferability of Perturbation

arXiv 2103.10274 ↗First posted 18 Mar 2021cs.LGcited by 13

Not on the map. Why?

Abstract (arXiv)

Deep neural networks (DNNs) are vulnerable to "backdoor" poisoning attacks, in which an adversary implants a secret trigger into an otherwise normally functioning model. Detection of backdoors in trained models without access to the training data or example triggers is an important open problem. In this paper, we identify an interesting property of these models: adversarial perturbations transfer from image to image more readily in poisoned models than in clean models. This holds for a variety of model and trigger types, including triggers that are not linearly separable from clean data. We use this feature to detect poisoned models in the TrojAI benchmark, as well as additional models.

Key facts. Cited by 13 papers in the map (data through 7 Oct 2026). Builds on 24 of its 40 references that are in the map. First posted on arXiv 18 Mar 2021.

Builds on

24 of 40 references.

References · 24

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,829
  4. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,283
  5. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,052
  6. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,012
  7. 2016End to End Learning for Self-Driving Carscited by 1,000
  8. 2016Universal adversarial perturbationscited by 774
  9. 2018Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networkscited by 550
  10. 2019STRIP: A Defence Against Trojan Attacks on Deep Neural Networkscited by 437
  11. 2018Detecting Backdoor Attacks on Deep Neural Networks by Activation Clusteringcited by 429
  12. 2020Backdoor Learning: A Surveycited by 412
12 more
  1. 2019Hidden Trigger Backdoor Attackscited by 359
  2. 2017Neural Trojanscited by 151
  3. 2018SentiNet: Detecting Localized Universal Attacks Against Deep Learning Systemscited by 148
  4. 2019TABOR: A Highly Accurate Approach to Inspecting and Restoring Trojan Backdoors in AI Systemscited by 120
  5. 2020Backdoor Attacks and Countermeasures on Deep Learning: A Comprehensive Reviewcited by 120
  6. 2021Backdoor Scanning for Deep Neural Networks through K-Arm Optimizationcited by 96
  7. 2020Practical Detection of Trojan Neural Networks: Data-Limited and Data-Free Casescited by 93
  8. 2020DeepSweep: An Evaluation Framework for Mitigating DNN Backdoor Attacks using Data Augmentationcited by 75
  9. 2019advertorch v0.1: An Adversarial Robustness Toolbox based on PyTorchcited by 69
  10. 2020The TrojAI Software Framework: An OpenSource tool for Embedding Trojans into Deep Learning Modelscited by 41
  11. 2020Poisoned classifiers are not only backdoored, they are fundamentally brokencited by 15
  12. 2021Baseline Pruning-Based Approach to Trojan Detection in Neural Networkscited by 5

Built on by

Cited by year

Citing papers per year: 2021: 4; 2022: 4; 2023: 3; 2024: 2.

13 papers.

Sort:

By area:

  1. 2021Trigger Hunting with a Topological Prior for Trojan Detectioncited by 39
  2. 2022A Survey of Neural Trojan Attacks and Defenses in Deep Learningcited by 17
  3. 2023UMD: Unsupervised Model Detection for X2X Backdoor Attackscited by 17
  4. 2024Preference Poisoning Attacks on Reward Model Learningcited by 14
  5. 2022Selective Amnesia: On Efficient, High-Fidelity and Blind Suppression of Backdoor Effects in Trojaned Machine Learning Modelscited by 11
  6. 2021Accumulative Poisoning Attacks on Real-time Datacited by 10
  7. 2023Single Image Backdoor Inversion via Robust Smoothed Classifierscited by 10
  8. 2021Check Your Other Door! Creating Backdoor Attacks in the Frequency Domaincited by 9
  9. 2023Don't FREAK Out: A Frequency-Inspired Approach to Detecting Backdoor Poisoned Samples in DNNscited by 8
  10. 2021MISA: Online Defense of Trojaned Models using Misattributionscited by 4
  11. 2022An Adaptive Black-box Backdoor Detection Method for Deep Neural Networkscited by 1
  12. 2022Dormant Neural Trojanscited by 0
1 more
  1. 2024Solving Trojan Detection Competitions with Linear Weight Classificationcited by 0