Abstract
Inverse reinforcement learning (IRL) is a useful tool for building autonomous agents capable of making decisions by learning from the behavioral records of human decision-makers. Incorporating individual differences into multi-agent systems can be key to solving complex problems that are difficult or impossible for an individual agent or a monolithic system to solve. In this paper, we propose a computational framework to predict decision-making competence (DMC) affecting cognitive ability to make decisions, using reward distributions uncovered by IRL. Our framework consists of building Double Transition Models (DTMs) from behavioral records of human subjects, discovering reward distributions using IRL, and utilizing these distributions to predict the DMC of individuals. The experimental results obtained by applying clustering and classification approaches confirmed that the DMC of the SCOUT subjects can be successfully predicted from reward distributions. High DMC individuals were clustered together and predicted more accurately than others, resulting in about 85 percent of predictive accuracy.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Proceedings of the Human Factors and Ergonomics Society Annual Meeting
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
