Theoretical Corrections and the Leveraging of Reinforcement Learning to Enhance Triangle Attack
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915025926488064 |
|---|---|
| author | Meng, Nicole Manicke, Caleb Chen, David Lao, Yingjie Ding, Caiwen Hong, Pengyu Mahmood, Kaleel |
| author_facet | Meng, Nicole Manicke, Caleb Chen, David Lao, Yingjie Ding, Caiwen Hong, Pengyu Mahmood, Kaleel |
| contents | Adversarial examples represent a serious issue for the application of machine learning models in many sensitive domains. For generating adversarial examples, decision based black-box attacks are one of the most practical techniques as they only require query access to the model. One of the most recently proposed state-of-the-art decision based black-box attacks is Triangle Attack (TA). In this paper, we offer a high-level description of TA and explain potential theoretical limitations. We then propose a new decision based black-box attack, Triangle Attack with Reinforcement Learning (TARL). Our new attack addresses the limits of TA by leveraging reinforcement learning. This creates an attack that can achieve similar, if not better, attack accuracy than TA with half as many queries on state-of-the-art classifiers and defenses across ImageNet and CIFAR-10. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2411_12071 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Theoretical Corrections and the Leveraging of Reinforcement Learning to Enhance Triangle Attack Meng, Nicole Manicke, Caleb Chen, David Lao, Yingjie Ding, Caiwen Hong, Pengyu Mahmood, Kaleel Machine Learning Cryptography and Security Adversarial examples represent a serious issue for the application of machine learning models in many sensitive domains. For generating adversarial examples, decision based black-box attacks are one of the most practical techniques as they only require query access to the model. One of the most recently proposed state-of-the-art decision based black-box attacks is Triangle Attack (TA). In this paper, we offer a high-level description of TA and explain potential theoretical limitations. We then propose a new decision based black-box attack, Triangle Attack with Reinforcement Learning (TARL). Our new attack addresses the limits of TA by leveraging reinforcement learning. This creates an attack that can achieve similar, if not better, attack accuracy than TA with half as many queries on state-of-the-art classifiers and defenses across ImageNet and CIFAR-10. |
| title | Theoretical Corrections and the Leveraging of Reinforcement Learning to Enhance Triangle Attack |
| topic | Machine Learning Cryptography and Security |
| url | https://arxiv.org/abs/2411.12071 |