Summary of Attaining Human`s Desirable Outcomes in Human-ai Interaction Via Structural Causal Games, by Anjie Liu et al.
Attaining Human`s Desirable Outcomes in Human-AI Interaction via Structural Causal Games
by Anjie Liu, Jianhong Wang, Haoxuan Li, Xu Chen, Jun Wang, Samuel Kaski, Mengyue Yang
First submitted to arxiv on: 26 May 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Computer Science and Game Theory (cs.GT); Human-Computer Interaction (cs.HC)
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 presents a theoretical framework for human-AI interaction, focusing on achieving the user’s desirable outcome through AI assistance. The challenge lies in navigating multiple Nash Equilibria that may not align with the desired outcome. To address this, the authors employ a structural causal game (SCG) framework and introduce a pre-policy intervention strategy to guide AI agents towards the desired outcome. A reinforcement learning-like algorithm is proposed to learn this pre-policy, which is tested in gridworld environments and realistic dialogue scenarios with large language models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In simple terms, this paper is about making sure that when humans use AI tools, they get what they want. Right now, it’s hard because there are many possible solutions, but not all of them work well for the human. The researchers created a new way to think about how humans and AI work together, using something called “structural causal games”. They also developed a new approach called “pre-policy intervention” that helps guide AI tools towards what the human wants. This was tested in different scenarios, including video games and talking to big language models. |
Keywords
* Artificial intelligence * Reinforcement learning
