A0324
Title: Integrating Logic Rule-Informed AI with Human Minds: Enhancing Collaborative Decision-Making
Authors: Yinghao Fu - Chinese university of hongkong, ShenZhen (China) [presenting]
Shuang Li - Chinese University of Hong Kong-Shenzhen (China)
Abstract: Integrating AI and human expertise is a promising approach to enhance decision-making in diverse scenarios. We present a novel framework to optimize decision-making by integrating and calibrating AI and human inputs. Human experts, despite their diverse backgrounds, may exhibit biased decision-making due to limitations in their expertise or cognitive domains, leading to suboptimal outcomes. To address this, we introduce an AI agent to complement the knowledge gaps of human experts, aiming to improve overall decision outcomes. Our AI agent, operating on probabilistic rule-based principles, provides informed decisions and works collaboratively with human experts. The decision-making process involves calibrating inputs from both AI and human experts to leverage collective insights for final decisions. Additionally, we adopt an online setting to facilitate continuous updates to the AI agent's knowledge base and refine calibration parameters, ensuring adaptability to evolving environments and new information. Experiments in simulation environments demonstrate that our model effectively integrates logic rule-informed AI with human expertise, enhancing collaborative decision-making.
