Summary of Approximating Human Models During Argumentation-based Dialogues, by Yinxu Tang et al.
Approximating Human Models During Argumentation-based Dialogues
by Yinxu Tang, Stylianos Loukas Vasileiou, William Yeoh
First submitted to arxiv on: 28 May 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- Secondary: Human-Computer Interaction (cs.HC); Logic in Computer Science (cs.LO)
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 A novel framework for Explainable AI Planning (XAIP) is proposed, enabling AI agents to learn and update a probabilistic human model through argumentation-based dialogues. The framework incorporates trust-based and certainty-based update mechanisms, allowing the agent to refine its understanding of the human’s mental state based on expressed trust in the agent’s arguments and certainty in their own arguments. The approach employs a probability weighting function inspired by prospect theory to capture the relationship between trust and perceived probability, and uses a Bayesian approach to update the agent’s probability distribution over possible human models. The framework is evaluated through a human-subject study, demonstrating its ability to capture the dynamics of human belief formation and adaptation. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary AI agents that can explain their decisions and actions to humans will help build trust between us and machines. One problem in making these agents is figuring out what’s going on inside people’s heads. Usually, we assume people think in a simple way, but this isn’t true for real-life interactions. This paper proposes a new method for AI agents to learn about human thought patterns by having conversations with humans. The AI adjusts its understanding of the human based on how much the human trusts the AI and how certain they are about their own thoughts. This helps the AI get better at guessing what people will do next. |
Keywords
» Artificial intelligence » Probability
