Enhancing Adversarial Example Transferability via Regularized Constrained Feature Layer
Source abstract
Abstract
Transfer-based Adversarial Attacks (TAAs) can deceive a victim model even without prior knowledge. This is achieved by leveraging the property of adversarial examples. That is, when generated from a surrogate model, they reta... | Find, read and cite all the research you need on Tech Science Press
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Fire Detection and Safety Systems · Adversarial Robustness in Machine Learning · Image Processing Techniques and Applications
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (28)
Show source-provided references
1. A comprehensive survey of robust deep learning in computer vision
2. A comprehensive study of ChatGPT: advancements, limitations, and ethical considerations in natural language processing and cybersecurity10.3390/info14080462 ↗
3. A systematic review of research on speech-recognition chatbots for language learning: implications for future directions in the era of large language models10.1080/10494820.2023.2204343 ↗
4. DETRs beat YOLOs on real-time object detection
5. A survey on adversarial attacks and defences10.1049/cit2.12028 ↗
6. Admix: enhancing the transferability of adversarial attacks
7. Transferable adversarial attack for both vision transformers and convolutional networks via momentum integrated gradients
8. GNP attack: transferable adversarial examples via gradient norm penalty
9. Structure invariant transformation for better adversarial transferability
10. Towards transferable adversarial attacks with centralized perturbation
11. An adaptive model ensemble adversarial attack for boosting adversarial transferability
12. Lgv: boosting adversarial example transferability from large geometric vicinity
13. Improving adversarial transferability via intermediate-level perturbation decay. Advances in neural information processing systems
14. Training meta-surrogate model for transferable adversarial attack
15. Transferable adversarial attacks on vision transformers with token gradient regularization
16. Toward understanding and boosting adversarial transferability from a distribution perspective10.1109/TIP.2022.3211736 ↗
17. Review on deep neural networks applied to low-frequency nilm10.3390/en14092390 ↗
18. Boosting transferability of adversarial samples via saliency distribution and frequency domain enhancement10.1016/j.knosys.2024.112152 ↗
19. Kurakin · Artificial intelligence safety and security · 2018 · 9910.1201/9781351251389-8 ↗
20. Boosting adversarial attacks with momentum
21. Improving transferability of adversarial examples with input diversity
22. Boosting adversarial transferability by achieving flat local maxima. Advances in neural information processing systems
23. Deep residual learning for image recognition
24. Densely connected convolutional networks
25. Going deeper with convolutions
26. Rethinking the inception architecture for computer vision
27. Iandola FN. SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5 MB model size. arXiv:1602.07360. 2016.
28. Tramèr F, Kurakin A, Papernot N, Goodfellow I, Boneh D, McDaniel P. Ensemble adversarial training: attacks and defenses. arXiv:1705.07204. 2017.