Summary of Adversarial Inception For Bounded Backdoor Poisoning in Deep Reinforcement Learning, by Ethan Rathbun et al.
Adversarial Inception for Bounded Backdoor Poisoning in Deep Reinforcement Learning
by Ethan Rathbun, Christopher Amato, Alina Oprea
First submitted to arxiv on: 17 Oct 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 The paper proposes a new class of backdoor attacks against Deep Reinforcement Learning (DRL) algorithms, dubbed “inception” attacks, that achieve state-of-the-art performance while minimizing alterations to the agent’s rewards. These attacks train agents to associate targeted adversarial behavior with high returns by inducing a disjunction between chosen and executed actions during training. The authors formally define these attacks and demonstrate their effectiveness in achieving both adversarial objectives. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper shows that DRL algorithms can be vulnerable to backdoor poisoning attacks, which cause agents to behave in certain ways when specific triggers are observed. Existing attacks rely on large changes to the agent’s rewards, making them detectable. The new “inception” attacks achieve better results with minimal reward changes. This means that DRL algorithms need to be improved to resist these types of attacks. |
Keywords
» Artificial intelligence » Reinforcement learning
