Together with Shihan Wang from Utrecht University, I will present SHARPIE at the European Workshop on Reinforcement Learning in Lille.
We will be presenting this paper as a poster. This is a systems paper presenting a system for controlled experiments involving human users and sequential decision-making agents that are common in e.g. reinforcement learning. We hope to connect with other hybrid human-RL researchers, meet potential users, and learn more about their research and how our system could support that in the future. What I find exciting about this work is that it helps understanding and shapes real human-agent interactions, and that it is currently already in use by some of our students and (professional) friends.