Summary of Theoretical Corrections and the Leveraging Of Reinforcement Learning to Enhance Triangle Attack, by Nicole Meng et al.
Theoretical Corrections and the Leveraging of Reinforcement Learning to Enhance Triangle Attack
by Nicole Meng, Caleb Manicke, David Chen, Yingjie Lao, Caiwen Ding, Pengyu Hong, Kaleel Mahmood
First submitted to arxiv on: 18 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Cryptography and Security (cs.CR)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel decision-based black-box attack called Triangle Attack with Reinforcement Learning (TARL), which addresses the limitations of the state-of-the-art Triangle Attack (TA). TARL leverages reinforcement learning to generate adversarial examples, achieving similar or better accuracy than TA with fewer queries on ImageNet and CIFAR-10 datasets. The paper provides a high-level description of TA and discusses its theoretical limitations, highlighting the importance of decision-based black-box attacks in sensitive domains. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The researchers develop a new attack method that can effectively generate adversarial examples for machine learning models. They build upon an existing technique called Triangle Attack (TA) and improve it by adding reinforcement learning. This makes their approach more efficient and accurate, requiring fewer queries to achieve similar results as TA. The paper shows the effectiveness of this new method on several datasets and models. |
Keywords
* Artificial intelligence * Machine learning * Reinforcement learning
