Scholarlink

With Human – Artificial Intelligence (AI) collaboration booming in all fields, the pace of task-based cooperation is ever-expanding. Yet, in most applications, AI induction is sidelined to test beds and is perceived skeptically as a competitor rather than a collaborator. The healthcare domain is one field where AI support is viewed as theoretical and far from practical. While most focus is directed towards developing and training AI models, the human expert and their interactions with the AI model are often overlooked. We present an experiment that incorporates, the personalization of human experts into the AI’s model training, aiming to improve collaboration and overall outcome. Using a simulation-based approach, we optimize the AI learning policy of a domain expert’s behaviour when evaluating decision support data of a patient’s risk of acquiring type 2 diabetes mellitus (T2DM). With Linear and Maximum Entropy inverse reinforcement learning (IRL) algorithms, we analyze various learning strategies by including context, rewards and sampling rates to show personalized expert characteristics with optimal policies and effective reward functions respectively. Our results provide insights into experts’ personalized evaluation policy and the AI model’s learning behaviour in various environmental scenarios and further the implicit difference in the evaluation of domain experts.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Simulation Modeling of Clinical Decision Making for Personalized Policy Identification

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Ashish T. S. Ireddy,
  • Sergey V. Kovalchuk

摘要

With Human – Artificial Intelligence (AI) collaboration booming in all fields, the pace of task-based cooperation is ever-expanding. Yet, in most applications, AI induction is sidelined to test beds and is perceived skeptically as a competitor rather than a collaborator. The healthcare domain is one field where AI support is viewed as theoretical and far from practical. While most focus is directed towards developing and training AI models, the human expert and their interactions with the AI model are often overlooked. We present an experiment that incorporates, the personalization of human experts into the AI’s model training, aiming to improve collaboration and overall outcome. Using a simulation-based approach, we optimize the AI learning policy of a domain expert’s behaviour when evaluating decision support data of a patient’s risk of acquiring type 2 diabetes mellitus (T2DM). With Linear and Maximum Entropy inverse reinforcement learning (IRL) algorithms, we analyze various learning strategies by including context, rewards and sampling rates to show personalized expert characteristics with optimal policies and effective reward functions respectively. Our results provide insights into experts’ personalized evaluation policy and the AI model’s learning behaviour in various environmental scenarios and further the implicit difference in the evaluation of domain experts.