Summary of Why the Agent Made That Decision: Explaining Deep Reinforcement Learning with Vision Masks, by Rui Zuo et al.
Why the Agent Made that Decision: Explaining Deep Reinforcement Learning with Vision Masks
by Rui Zuo, Zifan Wang, Simon Khan, Garrett Ethan Katz, Qinru Qiu
First submitted to arxiv on: 25 Nov 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Machine Learning (cs.LG)
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 introduces VisionMask, a standalone explanation model that identifies critical regions in deep reinforcement learning (DRL) agents’ visual inputs, enabling transparent decision-making. Existing methods require retraining the agent or perturbation-based techniques, which compromise performance and lack knowledge accumulation. VisionMask is trained end-to-end, self-supervised, and preserves the agent’s integrity. It outperforms existing methods on Super Mario Bros (SMB) and three Atari games, achieving 14.9% higher insertion accuracy and a 30.08% higher F1-Score in reproducing original actions from visual explanations. VisionMask has implications for safety-critical applications like medical diagnosis and military operations. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper makes deep reinforcement learning (DRL) more transparent, so people can trust the decisions made by AI agents. Currently, it’s hard to understand why these agents make certain choices. The authors created a new model called VisionMask, which shows what parts of an image are most important for making a decision. This helps us see how the agent is thinking and makes AI more trustworthy. The new model works well on several games and could be used in areas like medicine or the military. |
Keywords
* Artificial intelligence * F1 score * Reinforcement learning * Self supervised
