Faculty Job Seminar by LEE Min Hun

Towards Human-AI/Robot Collaborative Systems:
Improving the Practices of Physical Stroke Rehabilitation

Rapid advances in artificial intelligence (AI) have made it increasingly applicable to support human work (e.g. healthcare). However, the achievement of only accurate predictions of AI systems is not sufficient for their deployment in practice. If not carefully designed with stakeholders, AI systems can exacerbate user experience, and easily be abandoned. Instead, it is critical that these systems are designed to leverage the best of human ability, but also assist to overcome human limitations. In this talk, the speaker will introduce his work on creating two interactive hybrid intelligence systems that augment a machine learning model with human feedback in the context of physical stroke rehabilitation therapy: 1) human-AI collaborative decision making on rehabilitation assessment for therapists and 2) human-robot collaborative rehabilitation therapy for post-stroke survivors. In addition, he will also share insights from the design, development, and evaluation of collaborative systems on rehabilitation with therapists and post-stroke survivors. Last but not least, the speaker will discuss emerging and future directions for his research, exploring the core challenges of creating effective human-AI/robot collaborative systems.

About the Speaker

Min Hun Lee is a final year PhD candidate at Carnegie Mellon University. His research interests lie at the intersection of human-computer/robot interaction (HCI/HRI) and machine learning (ML). Specifically, he designs, develops, and evaluates human-centered ML systems to address societal problems. His thesis focuses on creating interactive systems to improve the practices of stroke rehabilitation (e.g. a decision support system for therapists and a robotic coaching system for post-stroke survivors).

He is a tenure-track faculty candidate for the Human-Machine Collaborative Systems, Human-Machine Interaction cluster.