Abstract

In socially assistive robotics, an important research area is the development of adaptation techniques and their effect on human-robot interaction. We present a meta-learning based policy gradient method for addressing the problem of adaptation in human-robot interaction and also investigate its role as a mechanism for trust modelling. By building an escape room scenario in mixed reality with a robot, we test our hypothesis that bi-directional trust can be influenced by different adaptation algorithms. We found that our proposed model increased the perceived trustworthiness of the robot and influenced the dynamics of gaining human’s trust. Additionally, participants evaluated that the robot perceived them as more trustworthy during the interactions with the meta-learning based adaptation compared to the previously studied statistical adaptation model.

Bio: (Alex) Yuan Gao received his Master’s degree in machine learning with a minor in mathematics from the University of Helsinki and is currently a PhD candidate at Uppsala University. He is interested in developing AI-driven robots that can think and feel like real humans (e.g. Ex Machina). In particular, he is interested in deep/reinforcement/neuro-based learning approaches for robotic perception, control, and physical modelling of the robot’s environment, which can help us to understand ourselves and build a unified learning structure for an adaptive, efficient and robust complex robotic system. Currently, he is working on projects that can fill the gap between deep reinforcement learning and social robotics.